RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation

CVPR 2017 Guosheng LinAnton MilanChunhua ShenIan Reid

Recently, very deep convolutional neural networks (CNNs) have shown outstanding performance in object recognition and have also been the first choice for dense classification problems such as semantic segmentation. However, repeated subsampling operations like pooling or convolution striding in deep CNNs lead to a significant decrease in the initial image resolution... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Semantic Segmentation ADE20K RefineNet Validation mIoU 40.7 # 14
Semantic Segmentation ADE20K val RefineNet (ResNet-152) mIoU 40.70% # 22
Semantic Segmentation ADE20K val RefineNet (ResNet-101) mIoU 40.20% # 23
Semantic Segmentation Cityscapes test RefineNet (ResNet-101) Mean IoU (class) 73.6% # 38
Semantic Segmentation NYU Depth v2 RefineNet (ResNet-101) Mean IoU 40.6% # 6
Semantic Segmentation PASCAL Context RefineNet mIoU 47.3 # 20
Semantic Segmentation PASCAL VOC 2012 test Multipath-RefineNet Mean IoU 84.2% # 15

Results from Other Papers


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK SOURCE PAPER COMPARE
Semantic Segmentation COCO-Stuff test RefineNet (ResNet-101) mIoU 33.6% # 7

Methods used in the Paper