Event-Based Vision: A Survey

Guillermo GallegoTobi DelbruckGarrick OrchardChiara BartolozziBrian TabaAndrea CensiStefan LeuteneggerAndrew DavisonJoerg ConradtKostas Daniilidis

article2019TPAMI2,835 citations

Presents a systematic overview of event-based vision by examining sensor hardware, processing algorithms from optical flow to 3D reconstruction, and neuromorphic computing methods for high-speed, high-dynamic-range perception.

Listen

Event cameras represent a fundamental shift from conventional frame-based imaging by asynchronously detecting per-pixel brightness changes and outputting sparse event streams that encode time, location, and polarity. Traditional cameras struggle with high-speed motion, extreme dynamic range, and low-latency requirements common in robotics and wearable systems, while event cameras offer microsecond temporal resolution, 140 dB dynamic range, low power consumption, and minimal motion blur.

The article surveys the emerging field of event-based vision to evaluate the technology's capabilities, document available sensors and processing methods, and identify pathways to practical deployment. It reviews the underlying pixel designs, event generation models, and both model-based and learning-based algorithms across low-level tasks such as feature detection, tracking, and optical flow estimation as well as higher-level tasks including 3D reconstruction, SLAM, motion segmentation, and object recognition.

Key findings show that event cameras can enable robust performance in scenarios where frame cameras fail, that motion-compensation and time-surface representations effectively aggregate sparse events for downstream processing, and that integration with neuromorphic processors and spiking networks yields low-power, low-latency systems. The survey also establishes that image reconstruction from events is feasible under appropriate regularization yet is often unnecessary when tasks are solved directly from the event stream.

These results matter because they demonstrate concrete routes to efficient, high-speed perception for autonomous vehicles, drones, AR/VR, and surveillance without the power or latency penalties of conventional cameras. Adoption could reduce system cost and energy use while expanding operational envelopes in uncontrolled lighting and fast motion.

Next steps include development of standardized benchmarks and larger annotated datasets, refinement of noise models and pixel miniaturization for mass production, and exploration of task-driven sensing that couples perception directly with control. Continued progress on hybrid event-frame pipelines and on-chip learning will further accelerate deployment.

The survey is comprehensive yet limited by the nascent state of the field, with many algorithms evaluated on small or synthetic datasets and hardware still relatively expensive. Confidence is high in the reported advantages and algorithmic trends, with moderate caution advised on quantitative performance claims until larger-scale, standardized evaluations become available.

ETH Zurich IBM Imperial College London Intel Italian Institute of Technology KTH Royal Institute of Technology Technische Universität Berlin University of Pennsylvania University of Zurich bandwidth computer vision low-power consumption event-based vision event-driven cameras event-based stereo matching

arXiv: 1904.08405 uzh-rpg/event-based_vision_resources

Paper: Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age, Cesar Cadena et al. (2016). This foundational survey outlines the core principles, state-estimation paradigms, and robustness challenges in visual SLAM that event-based vision systems directly aim to resolve in high-speed and high-dynamic-range settings.
Paper: FlowNet: Learning Optical Flow with Convolutional Networks, Philipp Fischer et al. (2015). Understanding this foundational work on learning optical flow from synthetic data provides key context for the learning-based motion and flow estimation techniques surveyed in event vision.
Paper: On Space-Time Interest Points, I. Laptev (2005). This classic paper establishes the mathematical formulation of space-time interest points, which underpins the spatiotemporal feature representation methods used for asynchronous event streams.
Paper: Efficient Processing of Deep Neural Networks: A Tutorial and Survey, Vivienne Sze et al. (2017). This tutorial details hardware accelerator dataflows and energy-efficiency bottlenecks in neural processing, providing necessary architectural context for deploying low-power event vision pipelines on neuromorphic hardware.
Paper: Unsupervised Learning of Depth and Ego-Motion from Video, Tinghui Zhou et al. (2017). Reading this paper provides the foundational photometric warping formulation for unsupervised joint learning of depth and camera ego-motion adapted by modern event-based visual odometry methods.
Paper: An efficient solution to the five-point relative pose problem, D. Nistér (2004). This paper provides the standard 5-point relative pose algorithm that serves as the baseline geometric solver integrated into event-based 3D reconstruction and visual odometry pipelines.

Paper: 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks, Christopher Choy et al. (2019). This work introduces 4D Minkowski sparse convolutional networks, providing an efficient spatio-temporal computational engine directly suitable for processing sparse, asynchronous event camera data.
Paper: RAFT: Recurrent All-Pairs Field Transforms for Optical Flow, Zachary Teed et al. (2020). This work develops the RAFT architecture for recurrent optical flow estimation, representing a major algorithmic paradigm that subsequent event-based motion estimation frameworks adapt and benchmark against.
Paper: D-NeRF: neural radiance fields for dynamic scenes, Albert Pumarola et al. (2021). This paper extends neural radiance fields to dynamic scenes using continuous temporal deformations, offering a continuous-time 3D reconstruction formulation complementary to high-temporal-resolution event imaging.

Abstract

Event cameras are bio-inspired sensors that differ from conventional frame cameras: Instead of capturing images at a fixed rate, they asynchronously measure per-pixel brightness changes, and output a stream of events that encode the time, location and sign of the brightness changes. Event cameras offer attractive properties compared to traditional cameras: high temporal resolution (in the order of microseconds), very high dynamic range (140 dB vs. 60 dB), low power consumption, and high pixel bandwidth (on the order of kHz) resulting in reduced motion blur. Hence, event cameras have a large potential for robotics and computer vision in challenging scenarios for traditional cameras, such as low-latency, high speed, and high dynamic range. However, novel methods are required to process the unconventional output of these sensors in order to unlock their potential. This paper provides a comprehensive overview of the emerging field of event-based vision, with a focus on the applications and the algorithms developed to unlock the outstanding properties of event cameras. We present event cameras from their working principle, the actual sensors that are available and the tasks that they have been used for, from low-level vision (feature detection and tracking, optic flow, etc.) to high-level vision (reconstruction, segmentation, recognition). We also discuss the techniques developed to process events, including learning-based techniques, as well as specialized processors for these novel sensors, such as spiking neural networks. Additionally, we highlight the challenges that remain to be tackled and the opportunities that lie ahead in the search for a more efficient, bio-inspired way for machines to perceive and interact with the world.

Topics

bandwidth computer vision low-power consumption motion blur reconstruction robotics sensors spiking neural networks

Keywords

event-based vision event-driven cameras event-based stereo matching event representations frame-based computer vision temporal contrast vision asynchronous vision visual odometry visual tracking

I Introduction and Applications
II Principle of Operation of Event Cameras
II-A Event Camera Designs
II-B Advantages of Event cameras
II-C Challenges Due To The Novel Sensing Paradigm
II-D Event Generation Model
II-E Event Camera Availability
III Event Processing
III-A Event Representations
III-B Methods for Event Processing
III-C Biologically Inspired Visual Processing
IV Algorithms / Applications
IV-A Feature Detection and Tracking
IV-B Optical Flow Estimation
IV-C 3D reconstruction. Monocular and Stereo
IV-D Pose Estimation and SLAM
IV-E Visual-Inertial Odometry (VIO)
IV-F Image Reconstruction
IV-G Motion Segmentation
IV-H Recognition
IV-I Neuromorphic Control
V Event-based Systems and Applications
V-A Neuromorphic Computing
V-B Applications in Real-Time On-Board Robotics
VI Resources
VI-A Software
VI-B Datasets and Simulators
VI-C Workshops
VII Discussion
VIII Conclusion
References

Knowls

Knowl 1 — Idealized Event Generation Model and Temporal Contrast Thresholding
equation
An event camera features an array of independent pixels that asynchronously respond to changes in their continuous logarithmic photocurrent (log intensity or brightness) $L(\mathbf{x}, t) \doteq \ln I(\mathbf{x}, t)$ , where $I(\mathbf{x}, t)$ is the pixel photocurrent at image coordinate $\mathbf{x} = (x, y)^\top$ and time $t$ .

In an idealized, noise-free operational model, an event $e_k = (\mathbf{x}_k, t_k, p_k)$ is asynchronously generated at pixel $\mathbf{x}_k = (x_k, y_k)^\top$ and timestamp $t_k$ whenever the temporal change in log intensity since the last event at that specific pixel exceeds a pre-set temporal contrast threshold $C > 0$ :

$\Delta L(\mathbf{x}_k, t_k) \doteq L(\mathbf{x}_k, t_k) - L(\mathbf{x}_k, t_k - \Delta t_k) = p_k C$

where $\Delta t_k$ is the time elapsed since the previous event occurred at the exact same pixel $\mathbf{x}_k$ , and $p_k \in \{+1, -1\}$ represents the binary polarity denoting whether brightness increased (ON event, $+1$ ) or decreased (OFF event, $-1$ ).

For small inter-event time intervals $\Delta t_k$ , first-order Taylor expansion yields $\Delta L(\mathbf{x}_k, t_k) \approx \frac{\partial L}{\partial t}(\mathbf{x}_k, t_k) \Delta t_k$ , which provides a direct relation between the discrete event stream and the continuous temporal derivative of brightness:

$\frac{\partial L}{\partial t}(\mathbf{x}_k, t_k) \approx \frac{p_k C}{\Delta t_k}$
Knowl 2 — Moving Edge Event Generation Model via Optical Flow Constraint
equation
Under the assumption of constant scene illumination and the classical brightness constancy assumption (optical flow constraint) $\frac{\partial L}{\partial t}(\mathbf{x}(t), t) + \nabla L(\mathbf{x}(t), t) \cdot \dot{\mathbf{x}}(t) = 0$ , the temporal intensity increment $\Delta L$ over a small time interval $\Delta t$ at an image point moving with velocity $\mathbf{v}(\mathbf{x}_k, t_k) = (v_x, v_y)^\top = \dot{\mathbf{x}}(t)$ is governed by:

$\Delta L \approx -\nabla L(\mathbf{x}_k, t_k) \cdot \mathbf{v}(\mathbf{x}_k, t_k) \Delta t$

where $\nabla L(\mathbf{x}_k, t_k) = (\partial_x L, \partial_y L)^\top$ is the spatial gradient of log intensity (representing scene edges) on the image plane, and $\Delta \mathbf{x} \doteq \mathbf{v} \Delta t$ is the spatial displacement.

This dot-product relationship imposes two geometric constraints on event generation:
1. If apparent motion is parallel to the scene edge ( $\mathbf{v} \cdot \nabla L = 0$ ), no brightness variation occurs and no events are generated.
2. If apparent motion is perpendicular to the scene edge ( $\mathbf{v} \parallel \nabla L$ ), brightness changes at the maximal rate, triggering events at the highest frequency.
Knowl 3 — Contrast Maximization and Image of Warped Events Framework
model/method
Contrast maximization is a unifying optimization framework for event-based vision tasks (including optical flow, ego-motion, depth estimation, and motion segmentation) operating on a spatio-temporal packet of events $\mathcal{E} = \{e_k\}_{k=1}^{N_e} = \{(\mathbf{x}_k, t_k, p_k)\}_{k=1}^{N_e}$ .

The method defines a continuous-time candidate motion model $\mathbf{W}(\mathbf{x}_k, t_k; \boldsymbol{ heta})$ that warps each event $\mathbf{x}_k$ along the motion hypothesis $\boldsymbol{ heta}$ to a chosen reference timestamp $t_{\text{ref}}$ :

$\mathbf{x}'_k = \mathbf{W}(\mathbf{x}_k, t_k; \boldsymbol{ heta})$

The warped events are accumulated onto a 2D spatial grid to construct an Image of Warped Events (IWE), $I(\mathbf{x}; \boldsymbol{ heta})$ :

$I(\mathbf{x}; \boldsymbol{ heta}) = \sum_{k=1}^{N_e} p_k \, \delta(\mathbf{x} - \mathbf{x}'_k)$

where $\delta$ is the 2D Dirac delta function (or a smooth approximation, such as a Gaussian or bilinear interpolation kernel, to permit continuous gradient descent).

When the candidate motion parameters $\boldsymbol{ heta}$ align with the true physical motion that generated the events, the warped trajectories overlap constructively, producing sharp, high-contrast edge patterns. The optimal motion parameters $\boldsymbol{ heta}^*$ are found by maximizing a focus objective function $g(I(\mathbf{x}; \boldsymbol{ heta}))$ , such as the image variance or the sum of squared spatial gradients:

$\boldsymbol{ heta}^* = \arg\max_{\boldsymbol{ heta}} g\left(I(\mathbf{x}; \boldsymbol{ heta})\right)$
Knowl 4 — Taxonomy of Event Data Representations
definition
Because asynchronous event streams cannot be directly ingested by standard frame-based computer vision routines, event processing algorithms aggregate and transform events into intermediate representations:
- Individual Events: $e_k = (\mathbf{x}_k, t_k, p_k)$ processed asynchronously one-by-one by filters (e.g., Kalman or particle filters) or Spiking Neural Networks (SNNs) to update an internal state with minimal latency.
- Event Packets: Spatio-temporal neighborhoods $\mathcal{E} = \{e_k\}_{k=1}^{N_e}$ containing a fixed number of events or events across a fixed temporal window $\Delta t$ .
- Event Frames / 2D Histograms: 2D grid representations formed by counting events or accumulating polarities per pixel over a time window: $H(\mathbf{x}) = \sum_{k=1}^{N_e} p_k \delta_{\mathbf{x}, \mathbf{x}_k}$ .
- Time Surfaces (TS) / Motion History Images: 2D maps storing a single temporal value per pixel, typically updated asynchronously with an exponential decay kernel emphasizing recent events: $T(\mathbf{x}, t) = \exp(-(t - t_{\text{last}}(\mathbf{x})) / \tau)$ , where $t_{\text{last}}(\mathbf{x})$ is the timestamp of the latest event at $\mathbf{x}$ and $\tau$ is a decay parameter.
- 3D Voxel Grids: Discretized space-time histograms $V(x, y, t_b)$ over $B$ discrete temporal bins where event polarities are bilinearly or linearly interpolated into adjacent temporal voxels, preserving temporal ordering and sub-voxel timing accuracy.
- 3D Point Sets: Point clouds $(\mathbf{x}_k, t_k) \in \mathbb{R}^3$ where time is treated as a third spatial coordinate for geometric fitting (such as local plane fitting) or PointNet architectures.
- Images of Warped Events (IWE): Motion-compensated 2D accumulations of events warped along candidate optical flow or camera velocity trajectories.
- Reconstructed Intensity Images: Continuous-time or high-frame-rate grayscale estimates derived by regularized temporal integration of event streams.

Knowl 5 — Characteristics and Specifications of Event-Based Vision Sensors

data/table

The following table compares key technical specifications across commercial and research event camera models:

Sensor Model	Resolution	Latency ( s)	Dynamic Range	Min. Contrast	Power (mW)	Pixel Size ( m)
DVS128 (2008)	$128 \times 128$	$12$	$120\text{ dB}$	$17\%$	$23$	$40 \times 40$
DAVIS240 (2014)	$240 \times 180$	$12$	$120\text{ dB}$	$11\%$	$5\text{--}14$	$18.5 \times 18.5$
DAVIS346 (2017)	$346 \times 260$	$20$	$120\text{ dB}$	$14.3\text{--}22.5\%$	$10\text{--}170$	$18.5 \times 18.5$
ATIS Gen3 (2011)	$304 \times 240$	$3$	$143\text{ dB}$	$13\%$	$50\text{--}175$	$30 \times 30$
Prophesee Gen4 CD (2020)	$1280 \times 720$	$20\text{--}150$	$>124\text{ dB}$	$11\%$	$32\text{--}84$	$4.86 \times 4.86$
Samsung DVS-Gen3 (2018)	$640 \times 480$	$50$	$90\text{ dB}$	$15\%$	$40$	$9 \times 9$
Samsung DVS-Gen4 (2020)	$1280 \times 960$	$150$	$100\text{ dB}$	$20\%$	$130$	$4.95 \times 4.95$
CeleX-V (2019)	$1280 \times 800$	$8$	$120\text{ dB}$	$10\%$	$400$	$9.8 \times 9.8$
Insightness Rino 3 (2018)	$320 \times 262$	$125$	$>100\text{ dB}$	$15\%$	$20\text{--}70$	$13 \times 13$

These specifications demonstrate the core operational envelope of event sensors: temporal latency in the microsecond regime ( $3\text{--}150\ \mu\text{s}$ ), very high dynamic range ( $90\text{--}143\text{ dB}$ compared to $\approx 60\text{ dB}$ for standard CMOS cameras), low power consumption (tens to hundreds of milliwatts), and a historical trend toward shrinking pixel pitch (from $40\ \mu\text{m}$ down to $4.86\ \mu\text{m}$ using 3D Back-Side Illuminated wafer stacking technology) and increasing spatial resolution (up to $1.2\text{ Mpixels}$ ).

Knowl 6 — Pixel Circuit Architectures: DVS, ATIS, and DAVIS
model/method
Event-based vision sensors rely on three primary pixel circuit topologies:
1. Dynamic Vision Sensor (DVS): Each pixel contains a continuous-time logarithmic photoreceptor coupled via a capacitor to a differential amplifier and two comparators with fixed voltage offsets ( $C^+, C^-$ ). When the change in log intensity exceeds the threshold, an asynchronous event is output over an Address-Event Representation (AER) bus, resetting the pixel state. DVS outputs purely binary polarity changes ( $p_k \in \{+1, -1\}$ ).
2. Asynchronous Time-Based Image Sensor (ATIS): Each pixel integrates two subpixels: a DVS change-detector (CD) subpixel and an exposure-measurement (EM) subpixel. The CD subpixel detects a brightness change and immediately triggers the EM subpixel, which measures absolute intensity by timing how quickly a dedicated photodiode discharges a capacitor between two threshold voltages. ATIS provides concurrent event triggering and high-dynamic-range ( $>120\text{ dB}$ ) asynchronous intensity measurement, but requires a larger pixel area.
3. Dynamic and Active Pixel Vision Sensor (DAVIS): Combines a DVS circuit and a standard Active Pixel Sensor (APS) readout circuit within every pixel, sharing a single photodiode. It provides concurrent asynchronous event streams (DVS) and synchronous global/rolling-shutter grayscale frames (APS, $\approx 55\text{ dB}$ dynamic range) while adding only $\approx 5\%$ area overhead to the DVS pixel.
Knowl 7 — Taxonomy of Event-Based Optical Flow Estimation Methods
model/method
Event-based optical flow algorithms compute image-plane velocities using several distinct paradigms, categorized along flow density (sparse vs. dense), flow component (normal flow perpendicular to local edges vs. full flow vectors), and architecture:
- Local Plane Fitting on Time Surfaces: Assumes the local space-time distribution of events forms a smooth surface. Fitting a local plane $t = a x + b y + c$ to the Time Surface yields the normal optical flow vector $\mathbf{v}_n = \frac{(a, b)^\top}{a^2 + b^2}$ . This method requires few events and minimal computation, but yields only the velocity component normal to the edge.
- Bio-Inspired Filter Banks and SNNs: Spatio-temporal Gabor filters, directional delay lines, and Spiking Neural Networks (SNNs) tuned via Spike-Timing-Dependent Plasticity (STDP) perform coincidence detection to extract velocity and orientation directly from spike timing.
- Variational Continuous Optimization: Minimizes energy functionals over 3D voxel grids combining the brightness constancy constraint with spatial/temporal smoothness regularizers, jointly outputting dense optical flow and reconstructed intensity.
- Contrast Maximization on Event Packets: Solves for flow parameters $\mathbf{v}$ by maximizing the edge sharpness (variance) of the Image of Warped Events (IWE) across local spatio-temporal patches.
- Deep Neural Networks: Encoder-decoder convolutional networks (e.g., EV-FlowNet) trained via self-supervised photometric loss across synchronized grayscale frames, or completely unsupervised by minimizing motion-compensation alignment loss on Time Surfaces or voxel grids.
Knowl 8 — Event-Based 3D Reconstruction and Depth Estimation Paradigms
model/method
Depth estimation using event cameras is divided into three distinct operational paradigms:
1. Instantaneous Multi-Camera Stereo: Employs two or more rigidly attached, time-synchronized event cameras. Events are matched across epipolar lines using temporal correlation (event simultaneity and concurrence of timestamps on Time Surfaces), normalized cross-correlation on local event frames, or cooperative spiking neural networks with mutual inhibition/excitation (implementing Marr-Poggio stereo principles) before triangulating 3D coordinates.
2. Monocular Space-Sweeping Depth (EMVS): Computes semi-dense 3D scene depth from a single moving event camera without requiring explicit data association or intensity reconstruction. Given known 6-DOF camera poses, every event back-projects into 3D space as a ray. Accumulating these rays across a discretized 3D projective voxel grid generates a ray density volume where true 3D scene edges manifest as sharp local maxima of ray density.
3. Active Structured Light Scanning: Pairs an event camera with an active emitter (such as a pulsed laser line or high-speed structured light pattern). By exploiting the sub-millisecond temporal resolution and redundancy suppression of the DVS, depth is extracted at high speeds and under extreme ambient lighting.

Knowl 9 — Complexity Axes and Taxonomy of Event-Based Pose Tracking and SLAM

data/table

Event-based SLAM and pose tracking algorithms are classified across three complexity dimensions: motion degrees of freedom (2D/3-DOF planar/rotational vs. 3D/6-DOF free motion), scene structure (artificial high-contrast planar/line patterns vs. natural 3D environments), and sensor requirements:

Method	Motion Dim	Tracking	Mapping/Depth	Scene Type	Events Only	Additional Sensors/Inputs
Cook et al. (2011)	2D	✓		Natural	✓	Rotational motion only
Weikersdorfer et al. (2013)	2D	✓		BW line	✓	Planar motion parallel to scene
Kim et al. (2014)	2D	✓		Natural	✓	Rotational motion only
Gallego Scaramuzza (2017)	2D	✓		Natural	✓	Rotational angular velocity
Censi Scaramuzza (2014)	3D	✓		BW		Attached depth sensor
Weikersdorfer et al. (2014)	3D	✓	✓	Natural		Attached RGB-D sensor
Mueggler et al. (2014)	3D	✓		BW	✓	Known 3D line map
Gallego et al. (2018)	3D	✓		Natural		Known 3D photometric map
Rebecq et al. (EMVS, 2018)	3D		✓	Natural	✓	Known camera poses
Kueng et al. (2016)	3D	✓	✓	Natural		Synchronous intensity frames
Kim et al. (2016)	3D	✓	✓	Natural	✓	Parallel filters + intensity recon.
Rebecq et al. (EVO, 2017)	3D	✓	✓	Natural	✓	None (Pure monocular event SLAM)

Pure event-based 6-DOF SLAM in natural 3D scenes without auxiliary sensors was achieved by interleaved probabilistic filtering (Kim et al., requiring GPU) and geometric parallel tracking and mapping (EVO, running real-time on CPU by pairing space-sweep mapping with edge-alignment pose tracking).

Knowl 10 — Visual-Inertial Odometry (VIO) Sensor Fusion Frameworks
model/method
To combine high-rate synchronous IMU inertial measurements (typically $\approx 1\text{ kHz}$ ) with asynchronous, spatially sparse event streams, event-based Visual-Inertial Odometry (VIO) utilizes three architectural strategies:
1. Asynchronous Filtering: Incoming events update an Extended Kalman Filter (EKF) or Multi-State Constraint Kalman Filter (MSCKF) state alongside IMU propagation steps, maintaining continuous real-time pose tracking.
2. Feature-Based Keyframe Optimization with IMU Pre-Integration: Features are detected and tracked asynchronously across motion-compensated event images or event frames. Feature coordinate tracks on the image plane are then fused with IMU readings using on-manifold IMU pre-integration in a keyframe-based nonlinear bundle adjustment optimization backend.
3. Continuous-Time Trajectory Optimization: The 6-DOF camera trajectory is modeled as a continuous parametric curve (e.g., cubic B-spline) along a unified temporal axis. An optimization objective function simultaneously minimizes continuous-time IMU inertial error residuals and event reprojection / photometric errors over trajectory segments without requiring explicit temporal discretization.
Knowl 11 — Event-Based Intensity Image Reconstruction and Frame Fusion
model/method
Because individual event sensors output only brightness changes $\Delta L$ , recovering absolute grayscale intensity images $L(\mathbf{x}, t)$ requires mathematical regularization:
- Gradient Estimation and Poisson Integration: Under rotational motion, temporal brightness derivatives $\partial L / \partial t$ are mapped to spatial intensity gradients $\nabla L$ using the optical flow constraint $\partial L / \partial t = -\nabla L \cdot \mathbf{v}$ . The gradient field is integrated across the image domain by solving the Poisson equation $\nabla^2 L = \nabla \cdot (\nabla L)$ .
- Variational and Manifold Regularization: Total variation (TV) or spatio-temporal smoothness penalties are minimized over discretized voxel grids or Time Surfaces to denoise and decompress events into video without explicit motion estimation.
- Learning-Based Recurrent Networks: Fully convolutional recurrent neural networks (e.g., UNet with ConvLSTM layers) convert interpolated voxel grids into high-frame-rate ( $2\text{--}5\text{ kHz}$ ) high-dynamic-range videos, leveraging perceptual priors learned from synthetic and natural video datasets.
- Event-Frame Complementary Filtering: Fuses low-frequency, motion-blurred or low-dynamic-range standard camera frames $F(\mathbf{x}, t)$ with high-frequency, low-latency event streams $E(\mathbf{x}, t)$ via temporal complementary filtering (low-pass on frames, high-pass on events) or double-integral models that simultaneously remove motion blur and restore high-speed HDR video.

Knowl 12 — Architectural Comparison of Neuromorphic Processors for Event Processing

data/table

Neuromorphic hardware architectures serve as parallel accelerators for Spiking Neural Networks (SNNs) driven by asynchronous event streams. Key specifications across primary processor families are summarized below:

Processor	SpiNNaker	TrueNorth	Loihi	DYNAP (Dynap-se)	Braindrop
Developer	U. Manchester	IBM	Intel	aiCTX	Stanford U.
Neuron Model Type	Software (ARM)	Digital	Digital	Analog	Analog
On-Chip Learning	Yes	No	Yes	No (Dynap-le: Yes)	No
CMOS Node	$130\text{ nm}$	$28\text{ nm}$	$14\text{ nm}$	$180\text{ nm}$	$28\text{ nm}$
Neurons per Chip	$4\text{k}$	$1024\text{k}$	$131\text{k}$	$1\text{k}$	$4\text{k}$
Synapses per Chip	$16\text{ M}$	$268\text{ M}$	$130\text{ M}$	$128\text{ k}$	$16\text{ M}$
Cores per Chip	$16$	$4096$	$128$	$4$	$1$
Software Stack	sPyNNaker, PACMAN	CPE/Eedn, NSCP	Nx SDK, Nengo	cAER, libcAER	Nengo

These processors span three implementation paradigms: (i) programmable software neurons executing on arrays of ARM cores (SpiNNaker) for maximum model flexibility, (ii) dedicated digital logic circuits (TrueNorth, Loihi) optimized for deterministic spiking throughput and on-chip STDP learning, and (iii) sub-threshold mixed-signal analog circuits (DYNAP, Braindrop) providing ultra-low power execution at the cost of analog device mismatch.

Knowl 13 — Motion Segmentation via Event Clustering and Contrast Maximization
model/method
When an event camera moves through a dynamic scene containing multiple independently moving objects, events are triggered simultaneously across the entire visual field by both the static background (due to camera ego-motion) and the moving objects. Motion segmentation resolves this by partitioning the event stream into discrete clusters corresponding to distinct motion models.

Formulated within the contrast maximization framework, the spatio-temporal event packet $\mathcal{E} = \{e_k\}_{k=1}^{N_e}$ is segmented into $M$ motion clusters with association weights $w_{k, m} \in [0, 1]$ satisfying $\sum_{m=1}^M w_{k, m} = 1$ . Each cluster $m$ is parameterized by motion parameters $\boldsymbol{ heta}_m$ (e.g., optical flow or planar homography).

The objective function maximizes the joint focus (contrast) across all $M$ motion-compensated event images:

$\max_{\{\boldsymbol{ heta}_m\}, \{w_{k,m}\}} \sum_{m=1}^M g\left( I_m(\mathbf{x}; \boldsymbol{ heta}_m, \{w_{k,m}\}) \right)$

where the $m$ -th Image of Warped Events is defined as:

$I_m(\mathbf{x}; \boldsymbol{ heta}_m, \{w_{k,m}\}) = \sum_{k=1}^{N_e} w_{k, m} \, p_k \, \delta\left(\mathbf{x} - \mathbf{W}(\mathbf{x}_k, t_k; \boldsymbol{ heta}_m)\right)$

Optimization alternates between updating motion parameters $\boldsymbol{ heta}_m$ via gradient ascent on the contrast metric and updating per-event cluster assignments $w_{k,m}$ .
Knowl 14 — Closed-Loop Neuromorphic Vision and Control Architectures
model/method
Event-based vision is utilized in closed-loop robotic perception-and-control systems across three architectural paradigms:
1. Neuromorphic-Vision-Driven Control: A neuromorphic sensor generates an event stream; an asynchronous event-based estimator computes state estimates (e.g., object position or vehicle pose); and a classical continuous or discrete controller asynchronously computes actuator control commands based on the updated state.
2. Native Neuromorphic Control: Bypasses explicit state estimation. Events are routed directly into spiking neural controllers or event-triggered control laws where individual sensory spikes or thresholded error events directly trigger actuator command changes, minimizing closed-loop latency to sub-millisecond timescales.
3. Task-Driven Bidirectional Perception-Control: Actuation and sensory processing are bidirectionally coupled. High-level task requirements dynamically modulate the low-level sensor configuration (e.g., adjusting on-chip analog bias currents, setting spatial regions of interest, or dynamically modifying refractory periods) to optimize the signal-to-noise ratio and bandwidth for the immediate task.

Coverage note — Specific benchmark classification accuracy numbers for individual datasets (e.g., N-MNIST, Poker-DVS) and specific robotic hardware testbed demonstrations (such as the pencil-balancing cart or robotic goalie) were omitted in favor of the foundational mathematical models, representations, algorithms, comparison tables, and system architectures.

References

1.M. Mahowald and C. Mead, “The silicon retina,” Scientific American, vol. 264, no. 5, pp. 76–83, May 1991.
2.P. Lichtsteiner, C. Posch, and T. Delbruck, “A 128×128 120 dB 15 µs latency asynchronous temporal contrast vision sensor,” IEEE J. Solid-State Circuits, vol. 43, no. 2, pp. 566–576, 2008.
3.C. Posch, D. Matolin, and R. Wohlgenannt, “A QVGA 143 dB dynamic range frame-free PWM image sensor with lossless pixellevel video compression and time-domain CDS,” IEEE J. Solid-State Circuits, vol. 46, no. 1, pp. 259–275, Jan. 2011.
4.C. Brandli, R. Berner, M. Yang, S.-C. Liu, and T. Delbruck, “A 240x180 130dB 3µs latency global shutter spatiotemporal vision sensor,” IEEE J. Solid-State Circuits, vol. 49, no. 10, pp. 2333–2341, 2014.
5.B. Son, Y. Suh, S. Kim, H. Jung, J.-S. Kim, C. Shin, K. Park, K. Lee, J. Park, J. Woo, Y. Roh, H. Lee, Y. Wang, I. Ovsiannikov, and H. Ryu, “A 640x480 dynamic vision sensor with a 9µm pixel and 300Meps address-event representation,” in IEEE Intl. Solid-State Circuits Conf. (ISSCC), 2017.
6.T. Delbruck and P. Lichtsteiner, “Fast sensory motor control based on event-based hybrid neuromorphic-procedural system,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2007, pp. 845–848.
7.G. Gallego, J. E. A. Lund, E. Mueggler, H. Rebecq, T. Delbruck, and D. Scaramuzza, “Event-based, 6-DOF camera tracking from photometric depth maps,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 40, no. 10, pp. 2402–2412, Oct. 2018.
8.H. Rebecq, R. Ranftl, V. Koltun, and D. Scaramuzza, “High speed and high dynamic range video with an event camera,” IEEE Trans. Pattern Anal. Mach. Intell., 2019.
9.https://github.com/uzh-rpg/event-based_vision_resources.
10.T. Delbruck, “Neuromorphic vision sensing and processing,” in Eur. Solid-State Device Research Conf. (ESSDERC), 2016, pp. 7–14.
11.S.-C. Liu, B. Rueckauer, E. Ceolini, A. Huber, and T. Delbruck, “Event-Driven Sensing for Efficient Perception: Vision and audition algorithms,” IEEE Signal Process. Mag., vol. 36, no. 6, pp. 29–37, Nov. 2019.
12.T. Delbruck and M. Lang, “Robotic goalie with 3ms reaction time at 4% CPU load using event-based dynamic vision sensor,” Front. Neurosci., vol. 7, p. 223, 2013.
13.A. Glover and C. Bartolozzi, “Event-driven ball detection and gaze fixation in clutter,” in IEEE Int. Conf. Intell. Robot. Syst. (IROS), 2016.
14.M. Litzenberger, A. N. Belbachir, N. Donath, G. Gritsch, H. Garn, B. Kohn, C. Posch, and S. Schraml, “Estimation of vehicle speed based on asynchronous data from a silicon retina optical sensor,” in IEEE Intell. Transp. Sys. Conf., 2006, pp. 653–658.
15.G. Orchard, C. Meyer, R. Etienne-Cummings, C. Posch, N. Thakor, and R. Benosman, “HFirst: A temporal approach to object recognition,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 37, no. 10, pp. 2028–2040, 2015.
16.J. H. Lee, T. Delbruck, M. Pfeiffer, P. K. Park, C.-W. Shin, H. Ryu, and B. C. Kang, “Real-time gesture interface based on eventdriven processing from stereo silicon retinas,” IEEE Trans. Neural Netw. Learn. Syst., vol. 25, no. 12, pp. 2250–2263, 2014.
17.A. Amir, B. Taba, D. Berg, T. Melano, J. McKinstry, C. D. Nolfo, T. Nayak, A. Andreopoulos, G. Garreau, M. Mendoza, J. Kusnitz, M. Debole, S. Esser, T. Delbruck, M. Flickner, and D. Modha, “A low power, fully event-based gesture recognition system,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2017.
18.P. Rogister, R. Benosman, S.-H. Ieng, P. Lichtsteiner, and T. Delbruck, “Asynchronous event-based binocular stereo matching,” IEEE Trans. Neural Netw. Learn. Syst., vol. 23, no. 2, pp. 347–353, 2012.
19.H. Rebecq, G. Gallego, E. Mueggler, and D. Scaramuzza, “EMVS: Event-based multi-view stereo—3D reconstruction with an event camera in real-time,” Int. J. Comput. Vis., vol. 126, no. 12, pp. 1394–1414, 2018.
20.N. Matsuda, O. Cossairt, and M. Gupta, “MC3D: Motion contrast 3D scanning,” in IEEE Int. Conf. Comput. Photography (ICCP), 2015.
21.R. Benosman, C. Clercq, X. Lagorce, S.-H. Ieng, and C. Bartolozzi, “Event-based visual flow,” IEEE Trans. Neural Netw. Learn. Syst., vol. 25, no. 2, pp. 407–417, 2014.
22.A. Z. Zhu, L. Yuan, K. Chaney, and K. Daniilidis, “EV-FlowNet: Self-supervised optical flow estimation for event-based cameras,” in Robotics: Science and Systems (RSS), 2018.
23.M. Cook, L. Gugelmann, F. Jug, C. Krautz, and A. Steger, “Interacting maps for fast visual interpretation,” in Int. Joint Conf. Neural Netw. (IJCNN), 2011, pp. 770–776.
24.H. Kim, A. Handa, R. Benosman, S.-H. Ieng, and A. J. Davison, “Simultaneous mosaicing and tracking with an event camera,” in British Mach. Vis. Conf. (BMVC), 2014.
25.H. Kim, S. Leutenegger, and A. J. Davison, “Real-time 3D reconstruction and 6-DoF tracking with an event camera,” in Eur. Conf. Comput. Vis. (ECCV), 2016.
26.H. Rebecq, T. Horstschafer, G. Gallego, and D. Scaramuzza, “EVO: A geometric approach to event-based 6-DOF parallel tracking and mapping in real-time,” IEEE Robot. Autom. Lett., vol. 2, no. 2, pp. 593–600, 2017.
27.A. Rosinol Vidal, H. Rebecq, T. Horstschaefer, and D. Scaramuzza, “Ultimate SLAM? combining events, images, and IMU for robust visual SLAM in HDR and high speed scenarios,” IEEE Robot. Autom. Lett., vol. 3, no. 2, pp. 994–1001, Apr. 2018.
28.L. Pan, C. Scheerlinck, X. Yu, R. Hartley, M. Liu, and Y. Dai, “Bringing a blurry frame alive at high frame-rate with an event camera,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2019.
29.G. Cohen, S. Afshar, A. van Schaik, A. Wabnitz, T. Bessell, M. Rutten, and B. Morreale, “Event-based sensing for space situational awareness,” in Proc. Advanced Maui Optical and Space Surveillance Technol. Conf. (AMOS), 2017.
30.T. Chin, S. Bagchi, A. P. Eriksson, and A. van Schaik, “Star tracking using an event camera,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2019.
31.P. Lichtsteiner and T. Delbruck, “64x64 event-driven logarithmic temporal derivative silicon retina,” in IEEE Workshop on ChargeCoupled Devices and Advanced Image Sensors, 2005, pp. 157–160.
32.——, “A 64x64 AER logarithmic temporal derivative silicon retina,” in Research in Microelectr. & Electron., PhD, vol. 2, 2005.
33.P. Lichtsteiner, “An AER temporal contrast vision sensor,” Ph. D. Thesis, ETH Zurich, Dept. of Physics (D-PHYS), Zurich, Switzerland, 2006.
34.P. Lichtsteiner, C. Posch, and T. Delbruck, “A 128x128 120dB 30mW asynchronous vision sensor that responds to relative intensity change,” in IEEE Intl. Solid-State Circuits Conf. (ISSCC), 2006, pp. 2060–2069.
35.D. Neil, “Deep neural networks and hardware systems for eventdriven data,” Ph.D. dissertation, ETH-Zurich, Zurich, Switzerland, 2017.
36.C. Posch, T. Serrano-Gotarredona, B. Linares-Barranco, and T. Delbruck, “Retinomorphic event-based vision sensors: Bioinspired cameras with spiking output,” Proc. IEEE, vol. 102, no. 10, pp. 1470–1484, Oct. 2014.
37.K. A. Boahen, “A burst-mode word-serial address-event link-I: Transmitter design,” IEEE Trans. Circuits Syst. I, vol. 51, no. 7, pp. 1269–1280, Jul. 2004.
38.S.-C. Liu, T. Delbruck, G. Indiveri, A. Whatley, and R. Douglas, Event-Based Neuromorphic Systems. John Wiley & Sons, 2015.
39.Y. Suh, S. Choi, M. Ito, J. Kim, Y. Lee, J. Seo, H. Jung, D.-H. Yeo, S. Namgung, J. Bong, J. seok Kim, P. K. J. Park, J. Kim, H. Ryu, and Y. Park, “A 1280x960 Dynamic Vision Sensor with a 4.95-µm pixel pitch and motion artifact minimization,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2020.
40.T. Delbruck, Y. Hu, and Z. He, “V2E: From video frames to realistic DVS event camera streams,” arXiv e-prints, 2020.
41.M. Mahowald, “VLSI analogs of neuronal visual processing: A synthesis of form and function,” Ph.D. dissertation, California Institute of Technology, Pasadena, California, May 1992.
42.D. il Dan Cho and T. jae Lee, “A review of bioinspired vision sensors and their applications,” Sens. Mater., vol. 27, no. 6, pp. 447–463, 2015.
43.L. Steffen, D. Reichard, J. Weinland, J. Kaiser, A. Ronnau, and R. Dillmann, “Neuromorphic stereo vision: A survey of bioinspired sensors and algorithms,” Front. Neurorobot., vol. 13, p. 28, 2019.
44.T. Delbruck and C. A. Mead, “Time-derivative adaptive silicon photoreceptor array,” in Proc. SPIE, Infrared sensors: Detectors, Electron., and Signal Process., vol. 1541, 1991, pp. 92–99.
45.S.-C. Liu and T. Delbruck, “Neuromorphic sensory systems,” Current Opinion in Neurobiology, vol. 20, no. 3, pp. 288–295, 2010.
46.T. Delbruck, B. Linares-Barranco, E. Culurciello, and C. Posch, “Activity-driven, event-based vision sensors,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2010, pp. 2426–2429.
47.T. Delbruck, “Fun with asynchronous vision sensors and processing,” in Eur. Conf. Comput. Vis. Workshops (ECCVW), 2012, pp. 506–515.
48.C. Posch, D. Matolin, and R. Wohlgenannt, “A QVGA 143dB dynamic range asynchronous address-event PWM dynamic image sensor with lossless pixel-level video compression,” in IEEE Intl. Solid-State Circuits Conf. (ISSCC), 2010, pp. 400–401.
49.E. Culurciello, R. Etienne-Cummings, and K. A. Boahen, “A biomorphic digital image sensor,” IEEE J. Solid-State Circuits, vol. 38, no. 2, pp. 281–294, Feb. 2003.
50.G. Orchard, D. Matolin, X. Lagorce, R. Benosman, and C. Posch, “Accelerated frame-free time-encoded multi-step imaging,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2014, pp. 2644–2647.
51.R. Berner, C. Brandli, M. Yang, S.-C. Liu, and T. Delbruck, “A 240x180 10mw 12µs latency sparse-output vision sensor for mobile applications,” in Proc. Symp. VLSI, 2013, pp. C186–C187.
52.E. Fossum, “CMOS image sensors: electronic camera-on-a-chip,” IEEE Trans. Electron Devices, vol. 44, no. 10, pp. 1689–1698, 1997.
53.R. Serrano-Gotarredona, M. Oster, P. Lichtsteiner, A. LinaresBarranco, R. Paz-Vicente, F. Gomez-Rodriguez, L. CamunasMesa, R. Berner, M. Rivas-Perez, T. Delbruck, S.-C. Liu, R. Douglas, P. Hafliger, G. Jimenez-Moreno, A. C. Ballcels, T. SerranoGotarredona, A. J. Acosta-Jimenez, and B. Linares-Barranco, “CAVIAR: A 45k neuron, 5M synapse, 12G connects/s AER hardware sensory-processing-learning-actuating system for highspeed visual object recognition and tracking,” IEEE Trans. Neural Netw., vol. 20, no. 9, pp. 1417–1438, 2009.
54.J. Conradt, M. Cook, R. Berner, P. Lichtsteiner, R. J. Douglas, and T. Delbruck, “A pencil balancing robot using a pair of AER dynamic vision sensors,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2009, pp. 781–784.
55.H. Xu, Y. Gao, F. Yu, and T. Darrell, “End-to-end learning of driving models from large-scale video datasets,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2017.
56.Y. Nozaki and T. Delbruck, “Temperature and parasitic photocurrent effects in dynamic vision sensors,” IEEE Trans. Electron Devices, vol. 64, no. 8, pp. 3239–3245, Aug. 2017.
57.——, “Authors Reply to Comment on Temperature and Parasitic Photocurrent Effects in Dynamic Vision Sensors,” IEEE Trans. Electron Devices, vol. 65, no. 7, pp. 3083–3083, Jul. 2018.
58.T. Serrano-Gotarredona and B. Linares-Barranco, “A 128x128 1.5% contrast sensitivity 0.9% FPN 3s latency 4mW asynchronous frame-free dynamic vision sensor using transimpedance preamplifiers,” IEEE J. Solid-State Circuits, vol. 48, no. 3, pp. 827–838, Mar. 2013.
59.M. Yang, S.-C. Liu, and T. Delbruck, “A dynamic vision sensor with 1% temporal contrast sensitivity and in-pixel asynchronous delta modulator for event encoding,” IEEE J. Solid-State Circuits, vol. 50, no. 9, pp. 2149–2160, 2015.
60.D. P. Moeys, F. Corradi, C. Li, S. A. Bamford, L. Longinotti, F. F. Voigt, S. Berry, G. Taverni, F. Helmchen, and T. Delbruck, “A sensitive dynamic and active pixel vision sensor for color or neural imaging applications,” IEEE Trans. Biomed. Circuits Syst., vol. 12, no. 1, pp. 123–136, Feb. 2018.
61.A. Rose, Vision: Human and Electronic. Plenum Press, New York, 1973.
62.C. Scheerlinck, N. Barnes, and R. Mahony, “Continuous-time intensity estimation using event cameras,” in Asian Conf. Comput. Vis. (ACCV), 2018.
63.——, “Asynchronous spatial image convolutions for event cameras,” IEEE Robot. Autom. Lett., vol. 4, no. 2, pp. 816–822, 2019.
64.D. Gehrig, H. Rebecq, G. Gallego, and D. Scaramuzza, “EKLT: Asynchronous photometric feature tracking using events and frames,” Int. J. Comput. Vis., 2019.
65.S. Bryner, G. Gallego, H. Rebecq, and D. Scaramuzza, “Eventbased, direct camera tracking from a photometric 3D map using nonlinear optimization,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2019.
66.G. Gallego, C. Forster, E. Mueggler, and D. Scaramuzza, “Eventbased camera pose tracking using a generative event model,” 2015, arXiv:1510.01972.
67.Prophesee Evaluation Kits, https://www.prophesee.ai/event-based-evk/, 2020.
68.T. Finateu, A. Niwa, D. Matolin, K. Tsuchimoto, A. Mascheroni, E. Reynaud, P. Mostafalu, F. Brady, L. Chotard, F. LeGoff, H. Takahashi, H. Wakabayashi, Y. Oike, and C. Posch, “A 1280x720 backilluminated stacked temporal contrast event-based vision sensor with 4.86µm pixels, 1.066geps readout, programmable event-rate controller and compressive data-formatting pipeline,” in IEEE Intl. Solid-State Circuits Conf. (ISSCC), 2020.
69.H. E. Ryu, “Industrial DVS design; key features and applications,” http://rpg.ifi.uzh.ch/docs/CVPR19workshop/CVPRW19_Eric_Ryu_Samsung.pdf.
70.M. Guo, J. Huang, and S. Chen, “Live demonstration: A 768 x 640 pixels 200meps dynamic vision sensor,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2017.
71.S. Chen and M. Guo, “Live demonstration: CeleX-V: A 1M pixel multi-mode event-based sensor,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2019.
72.Insightness Event-based Sensor Modules, http://www.insightness.com/technology/, 2020.
73.M. L. Katz, K. Nikolic, and T. Delbruck, “Live demonstration: Behavioural emulation of event-based vision sensors,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2012, pp. 736–740.
74.H. Rebecq, D. Gehrig, and D. Scaramuzza, “ESIM: an open event camera simulator,” in Conf. on Robotics Learning (CoRL), 2018.
75.A. Censi and D. Scaramuzza, “Low-latency event-based visual odometry,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2014.
76.M. Yang, S.-C. Liu, and T. Delbruck, “Analysis of encoding degradation in spiking sensors due to spike delay variation,” IEEE Trans. Circuits Syst. I, vol. 64, no. 1, pp. 145–155, Jan. 2017.
77.C. Posch and D. Matolin, “Sensitivity and uniformity of a 0.18µm CMOS temporal contrast pixel array,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2011, pp. 1572–1575.
78.C. Brandli, L. Muller, and T. Delbruck, “Real-time, high-speed video decompression using a frame- and event-based DAVIS sensor,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2014.
79.M. Yang, S.-C. Liu, and T. Delbruck, “Comparison of spike encoding schemes in asynchronous vision sensors: Modeling and design,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2014, pp. 2632–2635.
80.T. Delbruck, “Frame-free dynamic digital vision,” in Proc. Int. Symp. Secure-Life Electron., 2008, pp. 21–26.
81.A. Khodamoradi and R. Kastner, “O(N)-space spatiotemporal filter for reducing noise in neuromorphic vision sensors,” IEEE Trans. Emerg. Topics Comput., vol. PP, no. 99, pp. 1–1, 2018.
82.D. Czech and G. Orchard, “Evaluating noise filtering for eventbased asynchronous change detection image sensors,” in IEEE Int. Conf. Biomed. Robot. and Biomechatron. (BioRob), 2016, pp. 19–24.
83.V. Padala, A. Basu, and G. Orchard, “A noise filtering algorithm for event-based asynchronous change detection image sensors on TrueNorth and its implementation on TrueNorth,” Front. Neurosci., vol. 12, 2018.
84.T. Delbruck, V. Villanueva, and L. Longinotti, “Integration of dynamic vision sensor with inertial measurement unit for electronically stabilized event-based vision,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2014, pp. 2636–2639.
85.R. Berner, “Highspeed USB2.0 AER interfaces,” Master’s thesis, ETH Zurich, Dept. of Electrical and Information Eng. (D-ITET), Zurich, Switzerland, 2006.
86.iniVation, “Understandingthe performance of neuromorphic event-based vision sensors,” https://inivation.com/dvp/white-papers/, May 2020.
87.G. Taverni, D. P. Moeys, C. Li, C. Cavaco, V. Motsnyi, D. S. S. Bello, and T. Delbruck, “Front and back illuminated Dynamic and Active Pixel Vision Sensors comparison,” IEEE Trans. Circuits Syst. II, vol. 65, no. 5, pp. 677–681, 2018.
88.D. B. Fasnacht and T. Delbruck, “Dichromatic spectral measurement circuit in vanilla CMOS,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2007, pp. 3091–3094.
89.R. Berner, P. Lichtsteiner, and T. Delbruck, “Self-timed vertacolor dichromatic vision sensor for low power pattern detection,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2008, pp. 1032–1035.
90.L. Farian, J. A. Lenero-Bardallo, and P. Hafliger, “A bio-inspired AER temporal tri-color differentiator pixel array,” IEEE Trans. Biomed. Circuits Syst., vol. 9, no. 5, pp. 686–698, Oct. 2015.
91.R. B. Merrill, “Color separation in an active pixel cell imaging array using a triple-well structure,” US Patent US5 965 875A, 1999.
92.R. F. Lyon and P. M. Hubel, “Eyeing the camera: Into the next century,” in Proc. IS&T/SID 10th Color Imaging Conf., 2002, pp. 349–355.
93.C. Li, C. Brandli, R. Berner, H. Liu, M. Yang, S.-C. Liu, and T. Delbruck, “Design of an RGBW color VGA rolling and global shutter dynamic and active-pixel vision sensor,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2015.
94.D. P. Moeys, C. Li, J. N. P. Martel, S. Bamford, L. Longinotti, V. Motsnyi, D. S. S. Bello, and T. Delbruck, “Color temporal contrast sensitivity in dynamic vision sensors,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2017, pp. 1–4.
95.C. Scheerlinck, H. Rebecq, T. Stoffregen, N. Barnes, R. Mahony, and D. Scaramuzza, “CED: color event camera dataset,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2019.
96.A. Marcireau, S.-H. Ieng, C. Simon-Chane, and R. B. Benosman, “Event-based color segmentation with a high dynamic range sensor,” Front. Neurosci., vol. 12, 2018.
97.C. Posch, D. Matolin, R. Wohlgenannt, T. Maier, and M. Litzenberger, “A microbolometer asynchronous Dynamic Vision Sensor for LWIR,” IEEE Sensors J., vol. 9, no. 6, pp. 654–664, Jun. 2009.
98.E. Mueggler, H. Rebecq, G. Gallego, T. Delbruck, and D. Scaramuzza, “The event-camera dataset and simulator: Event-based data for pose estimation, visual odometry, and SLAM,” Int. J. Robot. Research, vol. 36, no. 2, pp. 142–149, 2017.
99.G. Gallego, M. Gehrig, and D. Scaramuzza, “Focus is all you need: Loss functions for event-based vision,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2019.
100.D. Gehrig, A. Loquercio, K. G. Derpanis, and D. Scaramuzza, “End-to-end learning of representations for asynchronous eventbased data,” in Int. Conf. Comput. Vis. (ICCV), 2019.
101.D. Weikersdorfer and J. Conradt, “Event-based particle filtering for robot self-localization,” in IEEE Int. Conf. Robot. Biomimetics (ROBIO), 2012.
102.F. Paredes-Valles, K. Y. W. Scheper, and G. C. H. E. de Croon, “Unsupervised learning of a hierarchical spiking neural network for optical flow estimation: From events to global motion perception,” IEEE Trans. Pattern Anal. Mach. Intell., Mar. 2019.
103.C. Reinbacher, G. Munda, and T. Pock, “Real-time panoramic tracking for event cameras,” in IEEE Int. Conf. Comput. Photography (ICCP), 2017, pp. 1–9.
104.E. Mueggler, G. Gallego, H. Rebecq, and D. Scaramuzza, “Continuous-time visual-inertial odometry for event cameras,” IEEE Trans. Robot., vol. 34, no. 6, pp. 1425–1440, Dec. 2018.
105.M. Liu and T. Delbruck, “Adaptive time-slice block-matching optical flow algorithm for dynamic vision sensors,” in British Mach. Vis. Conf. (BMVC), 2018.
106.A. Aimar, H. Mostafa, E. Calabrese, A. Rios-Navarro, R. Tapiador-Morales, I. Lungu, M. B. Milde, F. Corradi, A. Linares-Barranco, S. Liu, and T. Delbruck, “NullHop: A flexible convolutional neural network accelerator based on sparse representations of feature maps,” IEEE Trans. Neural Netw. Learn. Syst., vol. 30, no. 3, pp. 644–656, Mar. 2019.
107.J. Kogler, C. Sulzbachner, and W. Kubinger, “Bio-inspired stereo vision system with silicon retina imagers,” in Int. Conf. Comput. Vis. Syst. (ICVS), 2009, pp. 174–183.
108.A. I. Maqueda, A. Loquercio, G. Gallego, N. García, and D. Scaramuzza, “Event-based vision meets deep learning on steering prediction for self-driving cars,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2018.
109.X. Lagorce, G. Orchard, F. Gallupi, B. E. Shi, and R. Benosman, “HOTS: A hierarchy of event-based time-surfaces for pattern recognition,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 7, pp. 1346–1359, 2017.
110.M. A. R. Ahad, J. K. Tan, H. Kim, and S. Ishikawa, “Motion history image: its variants and applications,” Mach. Vis. and Applications, vol. 23, no. 2, pp. 255–281, Mar. 2012.
111.I. Alzugaray and M. Chli, “Asynchronous corner detection and tracking for event cameras in real time,” IEEE Robot. Autom. Lett., vol. 3, no. 4, pp. 3177–3184, Oct. 2018.
112.J. Manderscheid, A. Sironi, N. Bourdis, D. Migliore, and V. Lepetit, “Speed invariant time surface for learning to detect corner points with event-based cameras,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2019.
113.A. Sironi, M. Brambilla, N. Bourdis, X. Lagorce, and R. Benosman, “HATS: Histograms of averaged time surfaces for robust eventbased object classification,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2018.
114.V. Vasco, A. Glover, and C. Bartolozzi, “Fast event-based Harris corner detection exploiting the advantages of event-driven cameras,” in IEEE Int. Conf. Intell. Robot. Syst. (IROS), 2016.
115.E. Mueggler, C. Bartolozzi, and D. Scaramuzza, “Fast event-based corner detection,” in British Mach. Vis. Conf. (BMVC), 2017.
116.Y. Zhou, G. Gallego, H. Rebecq, L. Kneip, H. Li, and D. Scaramuzza, “Semi-dense 3D reconstruction with a stereo event camera,” in Eur. Conf. Comput. Vis. (ECCV), 2018.
117.P. Bardow, A. J. Davison, and S. Leutenegger, “Simultaneous optical flow and intensity estimation from an event camera,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2016.
118.S. Mostafavi I., L. Wang, Y.-S. Ho, and K.-J. Y. Yoon, “Eventbased high dynamic range image and very high frame rate video generation using conditional generative adversarial networks,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2019.
119.A. Z. Zhu, L. Yuan, K. Chaney, and K. Daniilidis, “Unsupervised event-based learning of optical flow, depth, and egomotion,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2019.
120.H. Rebecq, R. Ranftl, V. Koltun, and D. Scaramuzza, “Events-tovideo: Bringing modern computer vision to event cameras,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2019.
121.Y. Sekikawa, K. Hara, and H. Saito, “EventNet: Asynchronous recursive event processing,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2019.
122.M. Litzenberger, C. Posch, D. Bauer, A. N. Belbachir, P. Schon, B. Kohn, and H. Garn, “Embedded vision system for real-time object tracking using an asynchronous transient vision sensor,” in Digital Signal Processing Workshop, 2006, pp. 173–178.
123.Z. Ni, A. Bolopion, J. Agnus, R. Benosman, and S. Regnier, “Asynchronous event-based visual shape tracking for stable haptic feedback in microrobotics,” IEEE Trans. Robot., vol. 28, no. 5, pp. 1081–1089, 2012.
124.Z. Ni, S.-H. Ieng, C. Posch, S. Regnier, and R. Benosman, “Visual tracking using neuromorphic asynchronous event-based cameras,” Neural Computation, vol. 27, no. 4, pp. 925–953, 2015.
125.D. Tedaldi, G. Gallego, E. Mueggler, and D. Scaramuzza, “Feature detection and tracking with the dynamic and active-pixel vision sensor (DAVIS),” in Int. Conf. Event-Based Control, Comm. Signal Proc. (EBCCSP), 2016.
126.B. Kueng, E. Mueggler, G. Gallego, and D. Scaramuzza, “Lowlatency visual odometry using event-based feature tracks,” in IEEE Int. Conf. Intell. Robot. Syst. (IROS), 2016.
127.G. Gallego and D. Scaramuzza, “Accurate angular velocity estimation with an event camera,” IEEE Robot. Autom. Lett., vol. 2, no. 2, pp. 632–639, 2017.
128.G. Gallego, H. Rebecq, and D. Scaramuzza, “A unifying contrast maximization framework for event cameras, with applications to motion, depth, and optical flow estimation,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2018.
129.H. Rebecq, T. Horstschaefer, and D. Scaramuzza, “Real-time visual-inertial odometry for event cameras using keyframe-based nonlinear optimization,” in British Mach. Vis. Conf. (BMVC), 2017.
130.A. Z. Zhu, N. Atanasov, and K. Daniilidis, “Event-based feature tracking with probabilistic data association,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2017.
131.——, “Event-based visual inertial odometry,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2017.
132.A. Mitrokhin, C. Fermuller, C. Parameshwara, and Y. Aloimonos, “Event-based moving object detection and tracking,” in IEEE Int. Conf. Intell. Robot. Syst. (IROS), 2018.
133.A. Mitrokhin, C. Ye, C. Fermuller, Y. Aloimonos, and T. Delbruck, “EV-IMO: Motion segmentation dataset and learning pipeline for event cameras,” in IEEE Int. Conf. Intell. Robot. Syst. (IROS), 2019.
134.T. Brosch, S. Tschechne, and H. Neumann, “On event-based optical flow detection,” Front. Neurosci., vol. 9, Apr. 2015.
135.C. Reinbacher, G. Graber, and T. Pock, “Real-time intensity-image reconstruction for event cameras using manifold regularisation,” in British Mach. Vis. Conf. (BMVC), 2016.
136.H. Akolkar, S. Panzeri, and C. Bartolozzi, “Spike time based unsupervised learning of receptive fields for event-driven vision,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2015.
137.J. A. Perez-Carrasco, B. Zhao, C. Serrano, B. Acha, T. SerranoGotarredona, S. Chen, and B. Linares-Barranco, “Mapping from frame-driven to frame-free event-driven vision systems by lowrate rate coding and coincidence processing–application to feedforward ConvNets,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 11, pp. 2706–2719, Nov. 2013.
138.P. O’Connor, D. Neil, S.-C. Liu, T. Delbruck, and M. Pfeiffer, “Real-time classification and sensor fusion with a spiking deep belief network,” Front. Neurosci., vol. 7, p. 178, 2013.
139.P. U. Diehl, D. Neil, J. Binas, M. Cook, S.-C. Liu, and M. Pfeiffer, “Fast-classifying, high-accuracy spiking deep networks through weight and threshold balancing,” in Int. Joint Conf. Neural Netw. (IJCNN), vol. 4, 2015, pp. 2933–2940.
140.S. K. Esser, P. A. Merolla, J. V. Arthur, A. S. Cassidy, R. Appuswamy, A. Andreopoulos, D. J. Berg, J. L. McKinstry, T. Melano, D. R. Barch, C. di Nolfo, P. Datta, A. Amir, B. Taba, M. D. Flickner, and D. S. Modha, “Convolutional networks for fast, energy-efficient neuromorphic computing,” Proc. National Academy of Sciences, vol. 113, no. 41, pp. 11 441–11 446, 2016.
141.B. Rueckauer, I.-A. Lungu, Y. Hu, M. Pfeiffer, and S.-C. Liu, “Conversion of continuous-valued deep networks to efficient event-driven networks for image classification,” Front. Neurosci., vol. 11, p. 682, 2017.
142.S. B. Shrestha and G. Orchard, “SLAYER: Spike layer error reassignment in time,” in Conf. Neural Inf. Process. Syst. (NIPS), 2018.
143.J. H. Lee, T. Delbruck, and M. Pfeiffer, “Training deep spiking neural networks using backpropagation,” Front. Neurosci., vol. 10, p. 508, 2016.
144.E. Neftci, “Data and power efficient intelligence with neuromorphic learning machines,” iScience, 2018.
145.J. Kogler, C. Sulzbachner, M. Humenberger, and F. Eibensteiner, “Address-event based stereo vision with bio-inspired silicon retina imagers,” in Advances in Theory and Applications of Stereo Vision. InTech, 2011, pp. 165–188.
146.H. Li, G. Li, and L. Shi, “Classification of spatiotemporal events based on random forest,” in Advances in Brain Inspired Cognitive Systems (BICS), 2016.
147.A. Nguyen, T. Do, D. G. Caldwell, and N. G. Tsagarakis, “Realtime 6DOF pose relocalization for event cameras with stacked spatial LSTM networks,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2019.
148.E. Mueggler, C. Forster, N. Baumli, G. Gallego, and D. Scaramuzza, “Lifetime estimation of events from dynamic vision sensors,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2015.
149.S.-H. Ieng, J. Carneiro, M. Osswald, and R. Benosman, “Neuromorphic event-based generalized time-based stereovision,” Front. Neurosci., vol. 12, p. 442, 2018.
150.D. P. Moeys, F. Corradi, E. Kerr, P. Vance, G. Das, D. Neil, D. Kerr, and T. Delbruck, “Steering a predator robot using a mixed frame/event-driven convolutional neural network,” in Int. Conf. Event-Based Control, Comm. Signal Proc. (EBCCSP), 2016.
151.I.-A. Lungu, F. Corradi, and T. Delbruck, “Live demonstration: Convolutional neural network driven by dynamic vision sensor playing RoShamBo,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2017.
152.J. Binas, D. Neil, S.-C. Liu, and T. Delbruck, “DDD17: End-to-end DAVIS driving dataset,” in ICML Workshop on Machine Learning for Autonomous Vehicles, 2017.
153.C. Ye, A. Mitrokhin, C. Parameshwara, C. Fermuller, J. A. Yorke, and Y. Aloimonos, “Unsupervised learning of dense optical flow and depth from sparse event data,” arXiv e-prints, 2018, 1809.08625.
154.T. Stoffregen and L. Kleeman, “Simultaneous optical flow and segmentation (SOFAS) using Dynamic Vision Sensor,” in Australasian Conf. Robot. Autom. (ACRA), 2017.
155.T. Stoffregen, G. Gallego, T. Drummond, L. Kleeman, and D. Scaramuzza, “Event-based motion segmentation by motion compensation,” in Int. Conf. Comput. Vis. (ICCV), 2019.
156.L. A. Camunas-Mesa, T. Serrano-Gotarredona, and B. LinaresBarranco, “Event-driven sensing and processing for high-speed robotic vision,” in IEEE Biomed. Circuits Syst. Conf. (BioCAS), 2014.
157.G. Orchard, R. Benosman, R. Etienne-Cummings, and N. V. Thakor, “A spiking neural network architecture for visual motion estimation,” in IEEE Biomed. Circuits Syst. Conf. (BioCAS), 2013.
158.E. Chicca, P. Lichtsteiner, T. Delbruck, G. Indiveri, and R. Douglas, “Modeling orientation selectivity using a neuromorphic multi-chip system,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2006.
159.R. L. D. Valois, N. P. Cottaris, L. E. Mahon, S. D. Elfar, and J. Wilson, “Spatial and temporal receptive fields of geniculate and cortical cells and directional selectivity,” Vision Research, vol. 40, no. 27, pp. 3685–3702, 2000.
160.F. Rea, G. Metta, and C. Bartolozzi, “Event-driven visual attention for the humanoid robot iCub,” Front. Neurosci., vol. 7, 2013.
161.L. Itti and C. Koch, “Computational modelling of visual attention,” Nature Reviews Neuroscience, vol. 2, no. 3, pp. 194–203, 2001.
162.D. Marr and T. Poggio, “Cooperative computation of stereo disparity,” Science, vol. 194, no. 4262, pp. 283–287, 1976.
163.M. Mahowald, The Silicon Retina. Boston, MA: Springer US, 1994, pp. 4–65.
164.M. Osswald, S.-H. Ieng, R. Benosman, and G. Indiveri, “A spiking neural network model of 3D perception for event-based neuromorphic stereo vision systems,” Sci Rep, vol. 7, no. 1, Jan. 2017.
165.G. Dikov, M. Firouzi, F. Rohrbein, J. Conradt, and C. Richter, “Spiking cooperative stereo-matching at 2ms latency with neuromorphic hardware,” in Conf. Biomimetic and Biohybrid Systems, 2017, pp. 119–137.
166.E. Piatkowska, A. N. Belbachir, and M. Gelautz, “Asynchronous stereo vision for event-driven dynamic stereo sensor using an adaptive cooperative approach,” in Int. Conf. Comput. Vis. Workshops (ICCVW), 2013.
167.V. Vasco, A. Glover, Y. Tirupachuri, F. Solari, M. Chessa, and C. Bartolozzi, “Vergence control with a neuromorphic iCub,” in IEEE-RAS Int. Conf. Humanoid Robots (Humanoids), 2016.
168.M. Riesenhuber and T. Poggio, “Hierarchical models of object recognition in cortex,” Nature Neuroscience, vol. 2, no. 11, pp. 1019–1025, Nov. 1999.
169.H. Akolkar, C. Meyer, X. Clady, O. Marre, C. Bartolozzi, S. Panzeri, and R. Benosman, “What can neuromorphic event-driven precise timing add to spike-based pattern recognition?” Neural Computation, vol. 27, no. 3, pp. 561–593, Mar. 2015.
170.L. A. Camunas-Mesa, T. Serrano-Gotarredona, S. H. Ieng, R. B. Benosman, and B. Linares-Barranco, “On the use of orientation filters for 3D reconstruction in event-driven stereo vision,” Front. Neurosci., vol. 8, p. 48, 2014.
171.M. B. Milde, D. Neil, A. Aimar, T. Delbruck, and G. Indiveri, “ADaPTION: Toolbox and benchmark for training convolutional neural networks with reduced numerical precision weights and activation,” arXiv e-prints, Nov. 2017.
172.E. Stromatias, M. Soto, T. Serrano-Gotarredona, and B. LinaresBarranco, “An event-driven classifier for spiking neural networks fed with synthetic or dynamic vision sensor data,” Front. Neurosci., vol. 11, Jun. 2017.
173.E. Neftci, C. Augustine, S. Paul, and G. Detorakis, “Event-driven random back-propagation: Enabling neuromorphic deep learning machines,” Front. Neurosci., vol. 11, p. 324, 2017.
174.M. B. Milde, O. J. N. Bertrand, H. Ramachandran, M. Egelhaaf, and E. Chicca, “Spiking elementary motion detector in neuromorphic systems,” Neural Computation, vol. 30, no. 9, pp. 2384–2417, Sep. 2018.
175.H. Blum, A. Dietmller, M. Milde, J. Conradt, G. Indiveri, and Y. Sandamirskaya, “A neuromorphic controller for a robotic vehicle equipped with a dynamic vision sensor,” in Robotics: Science and Systems (RSS), 2017.
176.L. Salt, D. Howard, G. Indiveri, and Y. Sandamirskaya, “Differential evolution and bayesian optimisation for hyper-parameter selection in mixed-signal neuromorphic circuits applied to UAV obstacle avoidance,” arXiv e-prints, 2017, 1704.04853.
177.D. Drazen, P. Lichtsteiner, P. Hafliger, T. Delbrück, and A. Jensen, “Toward real-time particle tracking using an event-based dynamic vision sensor,” Experim. Fluids, vol. 51, no. 5, pp. 1465–1469, 2011.
178.Z. Ni, C. Pacoret, R. Benosman, S.-H. Ieng, and S. Regnier, “Asynchronous event-based high speed vision for microparticle tracking,” J. Microscopy, vol. 245, no. 3, pp. 236–244, 2012.
179.E. Piatkowska, A. N. Belbachir, S. Schraml, and M. Gelautz, “Spatiotemporal multiple persons tracking using dynamic vision sensor,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2012.
180.X. Lagorce, C. Meyer, S.-H. Ieng, D. Filliat, and R. Benosman, “Asynchronous event-based multikernel algorithm for highspeed visual features tracking,” IEEE Trans. Neural Netw. Learn. Syst., vol. 26, no. 8, pp. 1710–1720, 2015.
181.A. Glover and C. Bartolozzi, “Robust visual tracking with a freely-moving event camera,” in IEEE Int. Conf. Intell. Robot. Syst. (IROS), 2017.
182.D. R. Valeiras, X. Lagorce, X. Clady, C. Bartolozzi, S.-H. Ieng, and R. Benosman, “An asynchronous neuromorphic event-driven visual part-based shape tracking,” IEEE Trans. Neural Netw. Learn. Syst., vol. 26, no. 12, pp. 3045–3059, Dec. 2015.
183.M. A. Fischler and R. A. Elschlager, “The representation and matching of pictorial structures,” IEEE Trans. Comput., vol. C-22, no. 1, pp. 67–92, Jan. 1973.
184.C. Harris and M. Stephens, “A combined corner and edge detector,” in Proc. Fourth Alvey Vision Conf., vol. 15, 1988, pp. 147–151.
185.B. D. Lucas and T. Kanade, “An iterative image registration technique with an application to stereo vision,” in Int. Joint Conf. Artificial Intell. (IJCAI), 1981.
186.X. Clady, S.-H. Ieng, and R. Benosman, “Asynchronous eventbased corner detection and matching,” Neural Netw., vol. 66, pp. 91–106, 2015.
187.E. Rosten and T. Drummond, “Machine learning for high-speed corner detection,” in Eur. Conf. Comput. Vis. (ECCV), 2006.
188.V. Vasco, A. Glover, E. Mueggler, D. Scaramuzza, L. Natale, and C. Bartolozzi, “Independent motion detection with event-driven cameras,” in IEEE Int. Conf. Adv. Robot. (ICAR), 2017.
189.Y. Hu, H. Liu, M. Pfeiffer, and T. Delbruck, “DVS benchmark datasets for object tracking, action recognition, and object recognition,” Front. Neurosci., vol. 10, p. 405, 2016.
190.B. Rueckauer and T. Delbruck, “Evaluation of event-based algorithms for optical flow with ground-truth from inertial measurement sensor,” Front. Neurosci., vol. 10, no. 176, 2016.
191.R. Benosman, S.-H. Ieng, C. Clercq, C. Bartolozzi, and M. Srinivasan, “Asynchronous frameless event-based optical flow,” Neural Netw., vol. 27, pp. 32–37, 2012.
192.F. Barranco, C. Fermuller, and Y. Aloimonos, “Contour motion estimation for asynchronous event-driven cameras,” Proc. IEEE, vol. 102, no. 10, pp. 1537–1556, Oct. 2014.
193.——, “Bio-inspired motion estimation with event-driven sensors,” in Int. Work-Conf. Artificial Neural Netw. (IWANN), Advances in Comput. Intell., 2015, pp. 309–321.
194.J. Conradt, “On-board real-time optic-flow for miniature eventbased vision sensors,” in IEEE Int. Conf. Robot. Biomimetics (ROBIO), 2015.
195.S. Tschechne, R. Sailer, and H. Neumann, “Bio-inspired optic flow from event-based neuromorphic sensor input,” in Artificial Neural Networks in Pattern Recognition (ANNPR), 2014, pp. 171–182.
196.G. Haessig, A. Cassidy, R. Alvarez-Icaza, R. Benosman, and G. Orchard, “Spiking optical flow for event-based sensors using IBM’s truenorth neurosynaptic system,” IEEE Trans. Biomed. Circuits Syst., vol. 12, no. 4, pp. 860–870, Aug. 2018.
197.A. Z. Zhu, D. Thakur, T. Ozaslan, B. Pfrommer, V. Kumar, and K. Daniilidis, “The multivehicle stereo event camera dataset: An event camera dataset for 3D perception,” IEEE Robot. Autom. Lett., vol. 3, no. 3, pp. 2032–2039, Jul. 2018.
198.D. J. Fleet and A. D. Jepson, “Computation of component image velocity from local phase information,” Int. J. Comput. Vis., vol. 5, no. 1, pp. 77–104, 1990.
199.R. Hartley and A. Zisserman, Multiple View Geometry in Computer Vision. Cambridge University Press, 2003, 2nd Edition.
200.S. Schraml, A. N. Belbachir, N. Milosevic, and P. Schon, “Dynamic stereo vision system for real-time tracking,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2010, pp. 1409–1412.
201.J. Kogler, M. Humenberger, and C. Sulzbachner, “Event-based stereo matching approaches for frameless address event stereo data,” in Int. Symp. Adv. Vis. Comput. (ISVC), 2011, pp. 674–685.
202.J. Lee, T. Delbruck, P. K. J. Park, M. Pfeiffer, C.-W. Shin, H. Ryu, and B. C. Kang, “Live demonstration: Gesture-based remote control using stereo pair of dynamic vision sensors,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2012.
203.E. Piatkowska, A. N. Belbachir, and M. Gelautz, “Cooperative and asynchronous stereo vision for dynamic vision sensors,” Meas. Sci. Technol., vol. 25, no. 5, p. 055108, Apr. 2014.
204.R. Benosman, S.-H. Ieng, P. Rogister, and C. Posch, “Asynchronous event-based Hebbian epipolar geometry,” IEEE Trans. Neural Netw., vol. 22, no. 11, pp. 1723–1734, 2011.
205.J. Carneiro, S.-H. Ieng, C. Posch, and R. Benosman, “Event-based 3D reconstruction from neuromorphic retinas,” Neural Netw., vol. 45, pp. 27–38, 2013.
206.D. Zou, P. Guo, Q. Wang, X. Wang, G. Shao, F. Shi, J. Li, and P.-K. J. Park, “Context-aware event-driven stereo matching,” in IEEE Int. Conf. Image Process. (ICIP), 2016, pp. 1076–1080.
207.D. Zou, F. Shi, W. Liu, J. Li, Q. Wang, P.-K. J. Park, C.-W. Shi, Y. J. Roh, and H. E. Ryu, “Robust dense depth map estimation from sparse DVS stereos,” in British Mach. Vis. Conf. (BMVC), 2017.
208.M. Firouzi and J. Conradt, “Asynchronous event-based cooperative stereo matching using neuromorphic silicon retinas,” Neural Proc. Lett., vol. 43, no. 2, pp. 311–326, 2016.
209.J. Kogler, F. Eibensteiner, M. Humenberger, C. Sulzbachner, M. Gelautz, and J. Scharinger, “Enhancement of sparse silicon retina-based stereo matching using belief propagation and twostage postfiltering,” J. Electron. Imag., vol. 23, no. 4, pp. 1–15, 2014.
210.Z. Xie, S. Chen, and G. Orchard, “Event-based stereo depth estimation using belief propagation,” Front. Neurosci., vol. 11, 2017.
211.Z. Xie, J. Zhang, and P. Wang, “Event-based stereo matching using semiglobal matching,” Int. J. Advanced Robotic Systems, vol. 15, no. 1, 2018.
212.H. Hirschmuller, “Stereo processing by semiglobal matching and mutual information,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 30, no. 2, pp. 328–341, Feb. 2008.
213.A. Andreopoulos, H. J. Kashyap, T. K. Nayak, A. Amir, and M. D. Flickner, “A low power, high throughput, fully event-based stereo system,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2018.
214.A. Z. Zhu, Y. Chen, and K. Daniilidis, “Realtime time synchronized event-based stereo,” in Eur. Conf. Comput. Vis. (ECCV), 2018.
215.R. Szeliski, Computer Vision: Algorithms and Applications, ser. Texts in Computer Science. Springer, 2010.
216.S. Schraml, A. N. Belbachir, and H. Bischof, “An event-driven stereo system for real-time 3-D 360 panoramic vision,” IEEE Trans. Ind. Electron., vol. 63, no. 1, pp. 418–428, Jan. 2016.
217.——, “Event-driven stereo matching for real-time 3D panoramic vision,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2015.
218.C. Cadena, L. Carlone, H. Carrillo, Y. Latif, D. Scaramuzza, J. Neira, I. D. Reid, and J. J. Leonard, “Past, present, and future of simultaneous localization and mapping: Toward the robustperception age,” IEEE Trans. Robot., vol. 32, no. 6, pp. 1309–1332, 2016.
219.R. T. Collins, “A space-sweep approach to true multi-image matching,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 1996.
220.R. A. Newcombe, S. J. Lovegrove, and A. J. Davison, “DTAM: Dense tracking and mapping in real-time,” in Int. Conf. Comput. Vis. (ICCV), 2011.
221.C. Brandli, T. Mantel, M. Hutter, M. Hopflinger, R. Berner, R. Siegwart, and T. Delbruck, “Adaptive pulsed laser line extraction for terrain reconstruction using a dynamic vision sensor,” Front. Neurosci., vol. 7, p. 275, 2014.
222.J. N. P. Martel, J. Muller, J. Conradt, and Y. Sandamirskaya, “An active approach to solving the stereo matching problem using event-based sensors,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2018.
223.S. Thrun, W. Burgard, and D. Fox, Probabilistic Robotics. The MIT Press, Cambridge, MA, 2005.
224.D. Weikersdorfer, R. Hoffmann, and J. Conradt, “Simultaneous localization and mapping for event-based vision systems,” in Int. Conf. Comput. Vis. Syst. (ICVS), 2013.
225.D. Weikersdorfer, D. B. Adrian, D. Cremers, and J. Conradt, “Event-based 3D SLAM with a depth-augmented dynamic vision sensor,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2014.
226.E. Mueggler, B. Huber, and D. Scaramuzza, “Event-based, 6-DOF pose tracking for high-speed maneuvers,” in IEEE Int. Conf. Intell. Robot. Syst. (IROS), 2014, pp. 2761–2768.
227.M. Milford, H. Kim, S. Leutenegger, and A. Davison, “Towards visual SLAM with event-based cameras,” in The Problem of Mobile Sensors Workshop in conjunction with RSS, 2015.
228.E. Mueggler, G. Gallego, and D. Scaramuzza, “Continuous-time trajectory estimation for event-based vision sensors,” in Robotics: Science and Systems (RSS), 2015.
229.Y. Zhou, G. Gallego, and S. Shen, “Event-based stereo visual odometry,” arXiv preprint arXiv:2007.15548, 2020.
230.C. Forster, L. Carlone, F. Dellaert, and D. Scaramuzza, “Onmanifold preintegration for real-time visual-inertial odometry,” IEEE Trans. Robot., vol. 33, no. 1, pp. 1–21, 2017.
231.A. I. Mourikis and S. I. Roumeliotis, “A multi-state constraint Kalman filter for vision-aided inertial navigation,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2007.
232.S. Leutenegger, S. Lynen, M. Bosse, R. Siegwart, and P. Furgale, “Keyframe-based visual-inertial SLAM using nonlinear optimization,” Int. J. Robot. Research, 2015.
233.A. Patron-Perez, S. Lovegrove, and G. Sibley, “A spline-based trajectory representation for sensor fusion and rolling shutter cameras,” Int. J. Comput. Vis., vol. 113, no. 3, pp. 208–219, 2015.
234.L. von Stumberg, V. Usenko, and D. Cremers, “Direct sparse visual-inertial odometry using dynamic marginalization,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2018.
235.S. Barua, Y. Miyatani, and A. Veeraraghavan, “Direct face detection and video reconstruction from event cameras,” in IEEE Winter Conf. Appl. Comput. Vis. (WACV), 2016, pp. 1–9.
236.G. Munda, C. Reinbacher, and T. Pock, “Real-time intensityimage reconstruction for event cameras using manifold regularisation,” Int. J. Comput. Vis., vol. 126, no. 12, pp. 1381–1393, 2018.
237.C. Scheerlinck, H. Rebecq, D. Gehrig, N. Barnes, R. Mahony, and D. Scaramuzza, “Fast image reconstruction with an event camera,” in IEEE Winter Conf. Appl. Comput. Vis. (WACV), 2020.
238.A. N. Belbachir, S. Schraml, M. Mayerhofer, and M. Hofstaetter, “A novel HDR depth camera for real-time 3D 360-degree panoramic vision,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2014.
239.G. Wiesmann, S. Schraml, M. Litzenberger, A. N. Belbachir, M. Hofstatter, and C. Bartolozzi, “Event-driven embodied system for feature extraction and object recognition in robotic applications,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2012.
240.D. Gehrig, M. Gehrig, J. Hidalgo-Carrio, and D. Scaramuzza, “Video to Events: Recycling video datasets for event cameras,” in IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), 2020.
241.A. Zanardi, A. J. Aumiller, J. Zilly, A. Censi, and E. Frazzoli, “Cross-modal learning filters for rgb-neuromorphic wormhole learning,” in Proceedings of Robotics: Science and Systems, FreiburgimBreisgau, Germany, June 2019.
242.G. Orchard, A. Jayawant, G. K. Cohen, and N. Thakor, “Converting static image datasets to spiking neuromorphic datasets using saccades,” Front. Neurosci., vol. 9, p. 437, 2015.
243.D. Neil and S.-C. Liu, “Effective sensor fusion with event-based sensors and deep network architectures,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2016, pp. 2282–2285.
244.Y. Wu, L. Deng, G. Li, J. Zhu, and L. Shi, “Spatio-temporal backpropagation for training high-performance spiking neural networks,” Front. Neurosci., vol. 12, 2018.
245.A. Yousefzadeh, G. Orchard, T. Serrano-Gotarredona, and B. Linares-Barranco, “Active perception with dynamic vision sensors. minimum saccades with optimum recognition,” IEEE Trans. Biomed. Circuits Syst., vol. 12, no. 4, pp. 927–939, Aug. 2018.
246.X. Clady, J.-M. Maro, S. Barre, and R. B. Benosman, “A motionbased feature for event-based pattern recognition,” Front. Neurosci., vol. 10, Jan. 2017.
247.C. A. Sims, “Implications of rational inattention,” J. Monetary Economics, vol. 50, pp. 665–690, 2003.
248.M. Miskowicz, Event-Based Control and Signal Processing, ser. Embedded Systems. CRC Press, 2018.
249.P. Tabuada, “An introduction to event-triggered and selftriggered control,” in IEEE Conf. Decision Control (CDC), 2012.
250.K. J. Astrom, Event Based Control. Berlin, Heidelberg: Springer Berlin Heidelberg, 2008, pp. 127–147.
251.B. Wang and M. Fu, “Comparison of periodic and event-based sampling for linear state estimation,” IFAC Proc. Volumes (IFACPapersOnline), vol. 19, pp. 5508–5513, 2014.
252.A. Censi, E. Mueller, E. Frazzoli, and S. Soatto, “A powerperformance approach to comparing sensor families, with application to comparing neuromorphic to traditional vision sensors,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2015.
253.E. Mueller, A. Censi, and E. Frazzoli, “Low-latency heading feedback control with neuromorphic vision sensors using efficient approximated incremental inference,” in IEEE Conf. Decision Control (CDC), 2015.
254.P. Singh, S. Z. Yong, J. Gregoire, A. Censi, and E. Frazzoli, “Stabilization of linear continuous-time systems using neuromorphic vision sensors,” in IEEE Conf. Decision Control (CDC), 2016.
255.A. Censi, “Efficient neuromorphic optomotor heading regulation,” in IEEE Am. Control Conf. (ACC), 2015.
256.E. Mueller, A. Censi, and E. Frazzoli, “Efficient high speed signal estimation with neuromorphic vision sensors,” in Int. Conf. EventBased Control, Comm. Signal Proc. (EBCCSP), 2015.
257.D. Kahneman, Thinking, fast and slow. Farrar, Straus & Giroux, 2011.
258.S. B. Furber, D. R. Lester, L. A. Plana, J. D. Garside, E. Painkras, S. Temple, and A. D. Brown, “Overview of the SpiNNaker system architecture,” IEEE Trans. Comput., vol. 62, no. 12, pp. 2454–2467, 2013.
259.F. Akopyan, J. Sawada, A. Cassidy, R. Alvarez-Icaza, J. Arthur, P. Merolla, N. Imam, Y. Nakamura, P. Datta, G.-J. Nam, B. Taba, M. Beakes, B. Brezzo, J. B. Kuang, R. Manohar, W. P. Risk, B. Jackson, and D. S. Modha, “TrueNorth: Design and tool flow of a 65 mW 1 million neuron programmable neurosynaptic chip,” IEEE Trans. Comput.-Aided Design Integr. Circuits Syst., vol. 34, no. 10, pp. 1537–1557, 2015.
260.M. Davies, N. Srinivasa, T.-H. Lin, G. Chinya, Y. Cao, S. H. Choday, G. Dimou, P. Joshi, N. Imam, S. Jain et al., “Loihi: A neuromorphic manycore processor with on-chip learning,” IEEE Micro, vol. 38, no. 1, pp. 82–99, 2018.
261.S. Moradi, N. Qiao, F. Stefanini, and G. Indiveri, “A scalable multicore architecture with heterogeneous memory structures for dynamic neuromorphic asynchronous processors (DYNAPs),” IEEE Trans. Biomed. Circuits Syst., vol. 12, no. 1, pp. 106–122, 2018.
262.A. S. Neckar, “Braindrop: a mixed signal neuromorphic architecture with a dynamical systems-based programming model,” Ph.D. dissertation, Stanford University, Stanford, CA, Jun. 2018.
263.L. A. Camunas-Mesa, B. Linares-Barranco, and T. SerranoGotarredona, “Neuromorphic spiking neural networks and their memristor-CMOS hardware implementations,” Materials, vol. 12, no. 17, p. 2745, Aug. 2019.
264.B. Rajendran, A. Sebastian, M. Schmuker, N. Srinivasa, and E. Eleftheriou, “Low-power neuromorphic hardware for signal processing applications: A review of architectural and systemlevel design approaches,” IEEE Signal Process. Mag., vol. 36, no. 6, pp. 97–110, Nov. 2019.
265.S. B. Furber, F. Galluppi, S. Temple, and L. A. Plana, “The SpiNNaker Project,” Proc. IEEE, vol. 102, no. 5, pp. 652–665, 2014.
266.G. Haessig, F. Galluppi, X. Lagorce, and R. Benosman, “Neuromorphic networks on the SpiNNaker platform,” in IEEE Int. Conf. Artificial Intell. Circuits Syst. (AICAS), 2019.
267.C. Richter, F. Rohrbein, and J. Conradt, “Bio inspired optic flow detection using neuromorphic hardware,” in Bernstein Conf., 2014.
268.A. Glover, A. B. Stokes, S. Furber, and C. Bartolozzi, “ATIS + SpiNNaker: a fully event-based visual tracking demonstration,” in IEEE/RSJ Int. Conf. Intell. Robot. Syst. Workshops (IROSW), 2018.
269.T. Serrano-Gotarredona, B. Linares-Barranco, F. Galluppi, L. Plana, and S. Furber, “ConvNets experiments on SpiNNaker,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2015.
270.P. A. Merolla, J. V. Arthur, R. Alvarez-Icaza, A. S. Cassidy, J. Sawada, F. Akopyan, B. L. Jackson, N. Imam, C. Guo, Y. Nakamura et al., “A million spiking-neuron integrated circuit with a scalable communication network and interface,” Science, vol. 345, no. 6197, pp. 668–673, 2014.
271.P. Blouw, X. Choo, E. Hunsberger, and C. Eliasmith, “Benchmarking Keyword Spotting Efficiency on Neuromorphic Hardware,” arXiv e-prints, p. arXiv:1812.01739, Dec. 2018.
272.T. Bekolay, J. Bergstra, E. Hunsberger, T. DeWolf, T. Stewart, D. Rasmussen, X. Choo, A. Voelker, and C. Eliasmith, “Nengo: a Python tool for building large-scale functional brain models,” Front. Neuroinf., vol. 7, p. 48, 2014.
273.C. Eliasmith, T. C. Stewart, X. Choo, T. Bekolay, T. DeWolf, Y. Tang, and D. Rasmussen, “A large-scale model of the functioning brain,” Science, vol. 338, no. 6111, pp. 1202–1205, 2012.
274.N. Waniek, J. Biedermann, and J. Conradt, “Cooperative SLAM on small mobile robots,” in IEEE Int. Conf. Robot. Biomimetics (ROBIO), 2015.
275.B. J. P. Hordijk, K. Y. Scheper, and G. C. D. Croon, “Vertical landing for micro air vehicles using event-based optical flow,” J. Field Robot., vol. 35, no. 1, pp. 69–90, Jan. 2017.
276.jAER Project, https://github.com/SensorsINI/jaer, 2007.
277.T. Delbruck, M. Pfeiffer, R. Juston, G. Orchard, E. Muggler, A. Linares-Barranco, and M. W. Tilden, “Human vs. computer slot car racing using an event and frame-based DAVIS vision sensor,” in IEEE Int. Symp. Circuits Syst. (ISCAS), 2015.
278.M. Quigley, K. Conley, B. Gerkey, J. Faust, T. Foote, J. Leibs, R. Wheeler, and A. Y. Ng, “ROS: an open-source Robot Operating System,” in ICRA Workshop Open Source Softw., vol. 3, no. 2, 2009, p. 5.
279.A. Glover, V. Vasco, M. Iacono, and C. Bartolozzi, “The eventdriven software library for YARP-with algorithms and iCub applications,” Front. Robotics and AI, vol. 4, p. 73, 2018.
280.Y. Hu, J. Binas, D. Nei, S.-C. Liu, and T. Delbruck, “DDD20 Endto-End event camera driving dataset: Fusing frames and events with deep learning for improved steering prediction,” in IEEE Int. Conf. Intell. Transportation Systems (ITSC), 2020.
281.F. Barranco, C. Fermuller, Y. Aloimonos, and T. Delbruck, “A dataset for visual navigation with neuromorphic methods,” Front. Neurosci., vol. 10, p. 49, 2016.
282.J. Delmerico, T. Cieslewski, H. Rebecq, M. Faessler, and D. Scaramuzza, “Are we ready for autonomous drone racing? the UZHFPV drone racing dataset,” in IEEE Int. Conf. Robot. Autom. (ICRA), 2019.
283.C. Tan, S. Lallee, and G. Orchard, “Benchmarking neuromorphic vision: lessons learnt from computer vision,” Front. Neurosci., vol. 9, p. 374, 2015.
284.S. Miao, G. Chen, X. Ning, Y. Zi, K. Ren, Z. Bing, and A. Knoll, “Neuromorphic vision datasets for pedestrian detection, action recognition, and fall detection,” Front. Neurorobot., 2019.
285.E. Calabrese, G. Taverni, C. Awai Easthope, S. Skriabine, F. Corradi, L. Longinotti, K. Eng, and T. Delbruck, “DHP19: Dynamic vision sensor 3D human pose dataset,” in IEEE Conf. Comput. Vis. Pattern Recog. Workshops (CVPRW), 2019.
286.G. P. Garcia, P. Camilleri, Q. Liu, and S. B. Furber, “pyDVS: An extensible, real-time dynamic vision sensor emulator using offthe-shelf hardware,” in IEEE Symp. Series Comput. Intell. (SSCI), 2016.
287.S. J. Carey, A. Lopich, D. R. Barr, B. Wang, and P. Dudek, “A 100,000 fps vision sensor with embedded 535 GOPS/W 256x256 SIMD processor array,” in VLSI Circuits Symp., 2013.
288.J. N. P. Martel, “Unconventional processing with unconventional visual sensing: Parallel, distributed and event based vision algorithms & systems,” Ph.D. dissertation, ETH-Zurich, Zurich, Switzerland, 2019.

Citation

MLA

Gallego, G., et al. “Event-based Vision: A Survey”. University of Zurich, 2019, https://doi.org/10.5167/UZH-185139.

APA

Gallego, G., Delbruck, T., Orchard, G., Bartolozzi, C., Taba, B., Censi, A., Leutenegger, S., Davison, A. P., Conradt, J., Daniilidis, K., & Scaramuzza, D. (2019). Event-based Vision: A Survey. University of Zurich. https://doi.org/10.5167/UZH-185139

Chicago

Gallego, G., T. Delbruck, G. Orchard, et al. 2019. “Event-based Vision: A Survey”. University of Zurich, ahead of print. https://doi.org/10.5167/UZH-185139.

Harvard

Gallego, G. et al. (2019) “Event-based Vision: A Survey”, University of Zurich [Preprint]. Available at: https://doi.org/10.5167/UZH-185139.

Vancouver

1. Gallego G, Delbruck T, Orchard G, et al (2019) Event-based Vision: A Survey. University of Zurich. https://doi.org/10.5167/UZH-185139

BibTeX

@article{https://doi.org/10.5167/uzh-185139,
  doi = {10.5167/UZH-185139},
  url = {https://www.zora.uzh.ch/handle/20.500.14742/168561},
  author = {Gallego, Guillermo and Delbruck, Tobias and Orchard, Garrick and Bartolozzi, Chiara and Taba, Brian and Censi, Andrea and Leutenegger, Stefan and Davison, Andrew P and Conradt, Jörg and Daniilidis, Kostas and Scaramuzza, Davide},
  keywords = {570 Life sciences; biology},
  language = {en},
  title = {Event-based Vision: A Survey},
  publisher = {University of Zurich},
  year = {2019}
}

Metadata:DOI registry

Source Code

This paper has an official code repository available. Click below to access the source code.

View Repository

Access the Paper

This paper is available from its original source. Click below to access the PDF.

Open PDF

Event-Based Vision: A Survey

Abstract

Topics

Keywords

Table of Contents

Knowls

Knowl 1 — Idealized Event Generation Model and Temporal Contrast Thresholding

Knowl 2 — Moving Edge Event Generation Model via Optical Flow Constraint

Knowl 3 — Contrast Maximization and Image of Warped Events Framework

Knowl 4 — Taxonomy of Event Data Representations

Knowl 5 — Characteristics and Specifications of Event-Based Vision Sensors

Knowl 6 — Pixel Circuit Architectures: DVS, ATIS, and DAVIS

Knowl 7 — Taxonomy of Event-Based Optical Flow Estimation Methods

Knowl 8 — Event-Based 3D Reconstruction and Depth Estimation Paradigms

Knowl 9 — Complexity Axes and Taxonomy of Event-Based Pose Tracking and SLAM

Knowl 10 — Visual-Inertial Odometry (VIO) Sensor Fusion Frameworks

Knowl 11 — Event-Based Intensity Image Reconstruction and Frame Fusion

Knowl 12 — Architectural Comparison of Neuromorphic Processors for Event Processing

Knowl 13 — Motion Segmentation via Event Clustering and Contrast Maximization

Knowl 14 — Closed-Loop Neuromorphic Vision and Control Architectures

References

Citation

BibTeX

Source Code

Access the Paper