Papers
arxiv:2608.30281

SELECT: SELEctive Context Transfer for Class-Incremental Semantic Segmentation

Published on Aug 31
Authors:
,
,

Abstract

SELECT improves class-incremental semantic segmentation by selectively transferring context from similar past classes to initialize new classes, reducing forgetting and background shift.

Class-Incremental Semantic Segmentation (CISS) is fundamentally challenged by catastrophic forgetting and background shift, where learning new concepts degrades performance on previously seen classes. While existing methods attempt to balance stability (retaining old knowledge) and plasticity (learning new knowledge), they often fail to leverage prior knowledge effectively. These approaches typically rely on indiscriminate knowledge transfer or ambiguous initializations, which can dilute crucial semantic information. To overcome this limitation, we propose SELECT, a novel approach for Selective Context Transfer, which instead grounds each new class in a small set of semantically similar past classes. Its core is a Context Transfer Attention mechanism that aggregates the learned tokens from similar classes into a structured initialization for the new class. To ensure this transfer does not corrupt the borrowed representations, we add a controlled noise perturbation and a margin-based context-transfer loss that enforces separation between the new class token and its source tokens. Extensive experiments on Pascal VOC and ADE20K show that SELECT consistently outperforms prior work, achieving mIoU of 2.2% on VOC and 2.8% on ADE, providing an effective handle on the stability-plasticity dilemma. Code is available at https://github.com/avigupta2798/SELECT.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2608.30281
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 1

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2608.30281 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2608.30281 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.