Exploiting Spatial-Temporal Semantic Consistency for Video Scene Parsing

09/06/2021

∙

by   Xingjian He, et al.

∙

0

∙

share

Compared with image scene parsing, video scene parsing introduces temporal information, which can effectively improve the consistency and accuracy of prediction. In this paper, we propose a Spatial-Temporal Semantic Consistency method to capture class-exclusive context information. Specifically, we design a spatial-temporal consistency loss to constrain the semantic consistency in spatial and temporal dimensions. In addition, we adopt an pseudo-labeling strategy to enrich the training dataset. We obtain the scores of 59.84 58.85 And our method wins the 1st place on VSPW challenge at ICCV2021.