Controllable Relation Disentanglement for Few-Shot Class-Incremental Learning

arXiv:2403.11070v1 Announce Type: new
Abstract: In this paper, we propose to tackle Few-Shot Class-Incremental Learning (FSCIL) from a new perspective, i.e., relation disentanglement, which means enhancing FSCIL via disentangling spurious relation between categories. The challenge of disentangling spurious correlations lies in the poor controllability of FSCIL. On one hand, an FSCIL model is required to be trained in an incremental manner and thus it is very hard to directly control relationships between categories of different sessions. On the other hand, …

abstract arxiv challenge class correlations cs.cv few-shot incremental lies paper perspective type via