keyword

ICNet

ICNet, short for Image Cascade Network, is a convolutional neural network architecture designed for real-time semantic segmentation of high-resolution images. It addresses the trade-off between segmentation accuracy and computational speed by processing an input image across multiple resolution branches in a cascading framework. Lower-resolution branches pass through deeper network layers to efficiently extract broad contextual and semantic information, while higher-resolution branches pass through shallower layers to preserve fine spatial details and object boundaries. These multi-scale representations are progressively merged using cascade feature fusion modules with multi-level label supervision, allowing the model to perform high-quality, pixel-wise classification at high frame rates on standard hardware.

1 item

ICNet for Real-Time Semantic Segmentation on High-Resolution Images

Hengshuang Zhao, Xiaojuan Qi, Xiaoyong Shen, Jianping Shi, Jiaya Jia

OrganizationsSenseTimeTencentThe Chinese University of Hong Kong

Why you should read this

Introduces an image cascade network that combines multi-resolution branches with cascade feature fusion to deliver real-time, high-accuracy semantic segmentation on high-resolution images.

We focus on the challenging task of real-time semantic segmentation in this paper. It finds many practical applications and yet is with fundamental difficulty of reducing a large portion of computation for pixel-wise label inference. We propose an image cascade network (ICNet) that incorporates multi-resolution branches under proper label guidance to address this challenge. We provide in-depth analysis of our framework and introduce the cascade feature fusion unit to quickly achieve high-quality segmentation. Our system yields real-time inference on a single GPU card with decent quality results evaluated on challenging datasets like Cityscapes, CamVid and COCO-Stuff.

Source

https://link.springer.com/content/pdf/10.1007/978-3-030-01219-9_25.pdf

Added

2026-09-24