Divide and Conquer: Compositional Experts for Generalized Novel Class Discovery

Muli Yang, Yuehua Zhu, Jiaping Yu, Aming Wu, Cheng Deng; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 14268-14277

Abstract


In response to the explosively-increasing requirement of annotated data, Novel Class Discovery (NCD) has emerged as a promising alternative to automatically recognize unknown classes without any annotation. To this end, a model makes use of a base set to learn basic semantic discriminability that can be transferred to recognize novel classes. Most existing works handle the base and novel sets using separate objectives within a two-stage training paradigm. Despite showing competitive performance on novel classes, they fail to generalize to recognizing samples from both base and novel sets. In this paper, we focus on this generalized setting of NCD (GNCD), and propose to divide and conquer it with two groups of Compositional Experts (ComEx). Each group of experts is designed to characterize the whole dataset in a comprehensive yet complementary fashion. With their union, we can solve GNCD in an efficient end-to-end manner. We further look into the drawback in current NCD methods, and propose to strengthen ComEx with global-to-local and local-to-local regularization. ComEx is evaluated on four popular benchmarks, showing clear superiority towards the goal of GNCD.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Yang_2022_CVPR, author = {Yang, Muli and Zhu, Yuehua and Yu, Jiaping and Wu, Aming and Deng, Cheng}, title = {Divide and Conquer: Compositional Experts for Generalized Novel Class Discovery}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2022}, pages = {14268-14277} }