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. Our approach has achieved new state-of-the-art results 51.7% mIoU on PASCAL-Context, 85.9% mIoU on PASCAL VOC 2012.
|Task||Dataset||Model||Metric name||Metric value||Global rank||Compare|
|Semantic Segmentation||ADE20K||EncNet||Score||0.5567||# 1|
|Semantic Segmentation||PASCAL VOC 2012||EncNet||Mean IoU||85.9%||# 4|