SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2503.22136 · CVPR · 2025

Beyond Background Shift: Rethinking Instance Replay in Continual Semantic Segmentation

Wei Feng, Fanhua Shang, Hongying Liu, Fan Lyu, Liang Wan, Tingliang Feng, Hongmei Yin

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

instances with new images. By effectively resolving background shifts in both stored and new images, EIR alleviates catastrophic forgetting in the CSS task, thereby enhancing the model's capacity for CSS. Experimental results validate the efficacy of our approach, which significantly outperforms state-of-the-art CSS methods. The code is available at https://github.com/YikeYin97/EIR.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2503.22136")
get_code_for_paper("2503.22136")
have("2503.22136")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2503.22136.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2503.22136.svg)](https://syntology.ai/paper/2503.22136)

Connect an agent — have() is free.