List your company
ResearchFormerly WizWand

Semantic Segmentation on PASCAL Person-Part (test)

73.9mIoUSVCNet
55.760.42565.1569.875Oct 8, 2018Dec 2, 2018Jan 26, 2019Mar 23, 2019May 17, 2019Jul 11, 2019Sep 5, 2019
Updated 8mo ago

Evaluation results

MethodLinks
SVCNet2019.0973.9
RefineNet-152Number of Parameters=134M, Runtime (ms)=69.37 ± 0.78, Input Resolution=625 x 468, Multi-scale evaluation=true2018.1068.8
RefineNet2019.0968.6
DenseDecoder2019.0968.6
RefineNet-101Number of Parameters=118M, Runtime (ms)=60.25 ± 0.53, Input Resolution=625 x 4682018.1067.6
RefineNet-LW-152Number of Parameters=62M, Runtime (ms)=35.82 ± 0.23, Input Resolution=625 x 4682018.1067.6
RefineNet-LW-101Number of Parameters=46M, Runtime (ms)=27.16 ± 0.19, Input Resolution=625 x 4682018.1066.7
RefineNet-50Number of Parameters=99M, Runtime (ms)=54.18 ± 0.46, Input Resolution=625 x 4682018.1065.7
DeepLab-v2-CRFNumber of Parameters=44M, Multi-scale evaluation=true2018.1064.9
RefineNet-LW-50Number of Parameters=27M, Runtime (ms)=19.56 ± 0.29, Input Resolution=625 x 4682018.1064.9
DeepLab-V22019.0964.9
DeepLab2019.0962.8
Graph LSTM2019.0960.2
LG-LSTM2019.0958
HAZN2019.0957.5
Attention2019.0956.4