Abstract
Novel category discovery (NCD), which is a challenging and emerging task, aims to cluster unlabelled instances with knowledge information transferred from labelled ones. A majority of recent state-of-the-art methods leverage contrastive learning to model labelled and unlabelled data simultaneously. Nevertheless, they suffer from inaccurate and insufficient positive samples, which are detrimental to NCD and even its generalized class discovery (GCD) setting. To solve this problem, we propose positive-augmented contrastive learning (PACL), which can mine more positive samples and additional pseudo-positive samples, while augmenting the loss cost corresponding to these positive pairs. Consequently, PACL alleviates the imbalance between positive and negative pairs in contrastive learning, and facilitates the knowledge transfer for novel class discovery. In addition, we develop a general feature rectification approach based on PACL to rectify the representation learning achieved by existing NCD or GCD models. Extensive experiments on three datasets exhibit the necessity and effectiveness of our approach for both NCD and GCD tasks, without loss of generality.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Cheng, B., et al.: Panoptic-deeplab: a simple, strong, and fast baseline for bottom-up panoptic segmentation. In: CVPR, pp. 12475–12485 (2020)
Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: Imagenet: a large-scale hierarchical image database. In: CVPR, pp. 248–255 (2009)
Dosovitskiy, A., et al.: An image is worth 16x16 words: transformers for image recognition at scale. In: ICLR (2021)
Fei, Y., Zhao, Z., Yang, S., Zhao, B.: Xcon: learning with experts for fine-grained category discovery. In: BMVC (2022)
Fini, E., Sangineto, E., Lathuilière, S., Zhong, Z., Nabi, M., Ricci, E.: A unified objective for novel class discovery. In: ICCV, pp. 9284–9292 (2021)
Hadsell, R., Chopra, S., LeCun, Y.: Dimensionality reduction by learning an invariant mapping. In: CVPR, pp. 1735–1742 (2006)
Han, K., Rebuffi, S.A., Ehrhardt, S., Vedaldi, A., Zisserman, A.: Autonovel: Automatically discovering and learning novel visual categories. IEEE TPAMI 44(10), 6767–6781 (2021)
Han, K., Vedaldi, A., Zisserman, A.: Learning to discover novel visual categories via deep transfer clustering. In: ICCV, pp. 8401–8409 (2019)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR, pp. 770–778 (2016)
Jia, X., Han, K., Zhu, Y., Green, B.: Joint representation learning and novel category discovery on single-and multi-modal data. In: ICCV, pp. 610–619 (2021)
Kim, S., Kim, D., Cho, M., Kwak, S.: Embedding transfer with label relaxation for improved metric learning. In: CVPR, pp. 3967–3976 (2021)
Krause, J., Stark, M., Deng, J., Fei-Fei, L.: 3D object representations for fine-grained categorization. In: ICCV Workshop, pp. 554–561 (2013)
Krizhevsky, A., Hinton, G., et al.: Learning multiple layers of features from tiny images (2009)
LI, Z., Otholt, J., Dai, B., Hu, D., Meinel, C., Yang, H.: A closer look at novel class discovery from the labeled set. In: NeurIPS (2022)
Oliver, A., Odena, A., Raffel, C.A., Cubuk, E.D., Goodfellow, I.: Realistic evaluation of deep semi-supervised learning algorithms. In: NeurIPS (2018)
Oord, A.v.d., Li, Y., Vinyals, O.: Representation learning with contrastive predictive coding. arXiv:1807.03748 (2018)
Pu, N., Zhong, Z., Sebe, N.: Dynamic conceptional contrastive learning for generalized category discovery. In: CVPR (2023)
Sutskever, I., Martens, J., Dahl, G., Hinton, G.: On the importance of initialization and momentum in deep learning. In: ICML, pp. 1139–1147 (2013)
Tan, M., Pang, R., Le, Q.V.: Efficientdet: scalable and efficient object detection. In: CVPR, pp. 10781–10790 (2020)
Tarvainen, A., Valpola, H.: Mean teachers are better role models: weight-averaged consistency targets improve semi-supervised deep learning results. In: NeurIPS (2017)
Vaze, S., Han, K., Vedaldi, A., Zisserman, A.: Generalized category discovery. In: CVPR, pp. 7492–7501 (2022)
Wah, C., Branson, S., Welinder, P., Perona, P., Belongie, S.: The caltech-ucsd birds-200-2011 dataset (2011)
Wang, H., Zhu, Y., Adam, H., Yuille, A., Chen, L.C.: Max-deeplab: end-to-end panoptic segmentation with mask transformers. In: CVPR, pp. 5463–5474 (2021)
Wen, X., Zhao, B., Qi, X.: Parametric classification for generalized category discovery: a baseline study. arXiv:2211.11727 (2022)
Yang, M., Zhu, Y., Yu, J., Wu, A., Deng, C.: Divide and conquer: compositional experts for generalized novel class discovery. In: CVPR, pp. 14268–14277 (2022)
Zhong, Z., Fini, E., Roy, S., Luo, Z., Ricci, E., Sebe, N.: Neighborhood contrastive learning for novel class discovery. In: CVPR, pp. 10867–10875 (2021)
Zhong, Z., Zhu, L., Luo, Z., Li, S., Yang, Y., Sebe, N.: Openmix: reviving known knowledge for discovering novel visual categories in an open world. In: CVPR, pp. 9462–9470 (2021)
Acknowledgements
This work was supported in part by the NSF of China under Grant Nos. 62102061 and 62272083, and in part by the Liaoning Provincial NSF under Grant 2022-MS-137 and 2022-MS-128, and in part by the Fundamental Research Funds for the Central Universities under Grant DUT21RC(3)024.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Cai, Y., Pu, N., Jia, Q., Wang, W., Liu, Y. (2024). Broaden Your Positives: A General Rectification Approach for Novel Class Discovery. In: Liu, Q., et al. Pattern Recognition and Computer Vision. PRCV 2023. Lecture Notes in Computer Science, vol 14428. Springer, Singapore. https://doi.org/10.1007/978-981-99-8462-6_13
Download citation
DOI: https://doi.org/10.1007/978-981-99-8462-6_13
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-8461-9
Online ISBN: 978-981-99-8462-6
eBook Packages: Computer ScienceComputer Science (R0)