Dynamic Warping Network for Semantic Video Segmentation

Table 6

Comparison of state-of-the-art semantic video segmentation models on the Cityscapes test set.

Method	Source	mIoU %
Clockwork [34]	ECCV2016	66.4
PEARL [35]	ICCV2017	75.4
LLVSS [18]	CVPR2018	76.8
Accel [8]	CVPR2019	75.5
TDNet [19]	CVPR2020	79.9
ESVS [17]	ECCV2020	76.6
PSPNet [33]	CVPR2017	80.2
PSPNet + NetWarp [6]	ICCV2017	80.5
PSPNet + GRFP [15]	CVPR2018	80.6
PSPNet + EFC [13]	AAAI2020	81.0
PSPNet + ours		81.1
PSPNet + ours		81.8
DANet [5]	CVPR2019	81.5
DANet + ours		82.1

Methods trained using both fine and coarse sets are marked with “.” The bold values denote our method can achieve the best accuracy compared with other state-of-the-art methods.