Contrastive Mean-Shift Learning for Generalized Category Discovery

arXiv:2404.09451v1 Announce Type: new
Abstract: We address the problem of generalized category discovery (GCD) that aims to partition a partially labeled collection of images; only a small part of the collection is labeled and the total number of target classes is unknown. To address this generalized image clustering problem, we revisit the mean-shift algorithm, i.e., a classic, powerful technique for mode seeking, and incorporate it into a contrastive learning framework. The proposed method, dubbed Contrastive Mean-Shift (CMS) learning, trains an …

abstract arxiv clustering collection cs.cv discovery generalized image images mean part shift small total type