Location-aware Upsampling for Semantic Segmentation

13 Nov 2019Xiangyu HeZitao MoQiang ChenAnda ChengPeisong WangJian Cheng

Many successful learning targets such as minimizing dice loss and cross-entropy loss have enabled unprecedented breakthroughs in segmentation tasks. Beyond these semantic metrics, this paper aims to introduce location supervision into semantic segmentation... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Semantic Segmentation ADE20K LaU-regression-loss Validation mIoU 45.02 # 1
Semantic Segmentation ADE20K LaU-regression-loss Test Score 0.5632 # 2
Semantic Segmentation ADE20K LaU-offset-loss Validation mIoU 44.55 # 4
Semantic Segmentation ADE20K LaU-offset-loss Test Score 0.5641 # 1
Semantic Segmentation ADE20K val LaU-regression-loss mIoU 45.02 # 4
Semantic Segmentation PASCAL Context LaU-regression-loss (ResNet-101) mIoU 53.9 # 4