Geometry-Stable Class-Incremental Semantic Segmentation

doi:10.1109/tcsvt.2026.3743780

Geometry-Stable Class-Incremental Semantic Segmentation

2026-10-15

0

PRE

AI

GONG YIHONG (Yihong Gong)

DOI：10.1109/tcsvt.2026.3743780

Original

Original request for help

Save

Abstract

En 中文

Class-Incremental Semantic Segmentation (CISS) requires a model to continually learn to segment new classes without forgetting previously learned ones. Existing CISS methods attempt to preserve old knowledge by replaying or mimicking past predictions. However, they fail to stabilize the underlying feature geometry under shifting background semantics, leading to prototype drift and forgetting. To address this problem, we propose a geometry-stable CISS paradigm that imposes an Orthogonal Equiangular Tight Frame (O-ETF) classifier. Our O-ETF keeps all foreground prototypes on a simplex ETF while assigning the background a dedicated prototype strictly orthogonal to the foreground span. This decouples background supervision from foreground semantics, providing geometry-guided supervision that helps mitigate foreground-background interference under background semantic shift. Based on this geometry, we build a shared-encoder and task-specific-decoder framework, where lightweight LoRA modules are used to adapt each decoder branch to newly introduced classes. During inference, we further design a Bayesian product-of-experts (Bayes-PoE) fusion strategy, which lifts branch-level predictions to the accumulated global label space and aggregates them in the shared O-ETF semantic space. Extensive experiments on ADE20K and Pascal VOC 2012 demonstrate the effectiveness of our method.

Keywords:

Class-incremental learning

Semantic segmentation

Neural collapse theory

Geometric stability

Geometry-Stable Class-Incremental Semantic Segmentation

Abstract

Journal

Organization

Cited Papers