Context Encoding for Semantic Segmentation

CVPR 2018 Hang ZhangKristin DanaJianping ShiZhongyue ZhangXiaogang WangAmbrish TyagiAmit Agrawal

Recent work has made significant progress in improving spatial resolution for pixelwise labeling with Fully Convolutional Network (FCN) framework by employing Dilated/Atrous convolution, utilizing multi-scale features and refining boundaries. In this paper, we explore the impact of global contextual information in semantic segmentation by introducing the Context Encoding Module, which captures the semantic context of scenes and selectively highlights class-dependent featuremaps... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Semantic Segmentation ADE20K EncNet Validation mIoU 44.65 # 2
Semantic Segmentation ADE20K EncNet Test Score 0.5567 # 2
Semantic Segmentation PASCAL Context EncNet mIoU 51.7 # 4
Semantic Segmentation PASCAL VOC 2012 EncNet Mean IoU 85.9% # 4