Semantic Segmentation on VSPW Dataset through Aggregation of Transformer Models

TL;DR

This paper presents a video semantic segmentation method using aggregated Transformer models, achieving high accuracy and competitive ranking in a major ICCV challenge.

Contribution

The novel approach combines outputs of SWIN and VOLO Transformers for improved video scene parsing performance.

Findings

Achieved 57.3% mIoU on VSPW dataset

Ranked 3rd in ICCV 2021 Video Scene Parsing Challenge

Demonstrated effectiveness of Transformer aggregation

Abstract

Semantic segmentation is an important task in computer vision, from which some important usage scenarios are derived, such as autonomous driving, scene parsing, etc. Due to the emphasis on the task of video semantic segmentation, we participated in this competition. In this report, we briefly introduce the solutions of team 'BetterThing' for the ICCV2021 - Video Scene Parsing in the Wild Challenge. Transformer is used as the backbone for extracting video frame features, and the final result is the aggregation of the output of two Transformer models, SWIN and VOLO. This solution achieves 57.3% mIoU, which is ranked 3rd place in the Video Scene Parsing in the Wild Challenge.