Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

high-resolution images

High-resolution images are digital visual representations that contain a large number of pixels and high spatial detail, allowing visual content to appear sharp, clear, and rich in fine structural information. In digital imaging and computer vision, these images feature expansive pixel dimensions across their height and width, capturing distinct object boundaries, fine textures, and subtle visual cues that would otherwise be degraded or lost in lower-resolution formats. While this dense pixel distribution is essential for tasks requiring precise spatial fidelity, such as automated visual recognition, remote sensing, and pixel-level scene segmentation, the increased data volume also demands substantially higher computational power, memory bandwidth, and storage capacity compared to standard-resolution images.

1 item

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

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.

Added

2026-09-24