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

Hengshuang Zhao1    Xiaojuan Qi1    Xiaoyong Shen2    Jianping Shi3    Jiaya Jia1,2   

1 The Chinese Univeristy of Hong Kong     2 Tencent Youtu Lab     3 SenseTime Research    

Network architecture of ICNet. ‘CFF’ stands for cascade feature fusion detailed in Sec. 3.3. Numbers in parentheses are feature map size ratios to the full-resolution input. Operations are highlighted in brackets. The final ×4 upsampling in the bottom branch is only used during testing.

Abstract

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.

Download

Snapshot for paper "ICNet for Real-Time Semantic Segmentation on High-Resolution Images"
Hengshuang Zhao, Xiaojuan Qi, Xiaoyong Shen, Jianping Shi, Jiaya Jia.
European Conference on Computer Vision (ECCV), 2018.

  [Paper]  [Supp]  [Bib]

  [Code]

Comparison

Comparison of semantic segmentation frameworks. (a) Intermediate skip connection used by FCN [1] and Hypercolumns [21]. (b) Encoder-decoder structure incorporated in SegNet [3], DeconvNet [4], UNet [33], ENet [8], and step-wise reconstruction & refinement from LRR [34] and RefineNet [11]. (c) Multi-scale prediction ensemble adopted by DeepLab-MSC [2] and PSPNet-MSC [5]. (d) Our ICNet architecture.

Performance

Inference speed and mIoU performance on Cityscapes [7] test set. Methods involved are PSPNet [5], ResNet38 [6], DUC [10], RefineNet [11], FRRN [12], DeepLabv2-CRF[13], Dilation10 [14], DPN [15], FCN-8s [1], DeepLab [2], CRF-RNN [16], SQ [9], ENet [8], SegNet [3], and our ICNet.

Time spent on PSPNet50 with dilation 8 for two input images. Roughly running time is proportional to the pixel number and kernel number.

Visualization

Visual prediction improvement of ICNet in each branch on Cityscapes dataset.

Visual prediction improvement of ICNet. White regions in 'diff1' and 'diff2' denote prediction difference between 'sub24' and 'sub4', and between 'sub124' and 'sub24' respectively.

[More Visualization]

Video

Demo video processed by ICNet on cityscapes dataset with TitanX Maxwell:

Last update: Aug. 18, 2018