Memory Based Video Scene Parsing

TL;DR

This paper presents a memory-based approach for video scene parsing, leveraging temporal information to improve pixel-wise semantic labeling across video frames, achieving competitive results in a challenging benchmark.

Contribution

The paper introduces a memory-based method specifically designed for video scene parsing, addressing the challenge of utilizing temporal data effectively.

Findings

Achieved a mIoU of 57.44 on the challenge dataset

Secured 2nd place in the Video Scene Parsing in the Wild Challenge

Demonstrated the effectiveness of memory-based techniques in video semantic segmentation

Abstract

Video scene parsing is a long-standing challenging task in computer vision, aiming to assign pre-defined semantic labels to pixels of all frames in a given video. Compared with image semantic segmentation, this task pays more attention on studying how to adopt the temporal information to obtain higher predictive accuracy. In this report, we introduce our solution for the 1st Video Scene Parsing in the Wild Challenge, which achieves a mIoU of 57.44 and obtained the 2nd place (our team name is CharlesBLWX).