Skip to main content
placeholder image

Contrastive Positive Mining for��Unsupervised 3D Action Representation Learning

Chapter


Abstract


  • Recent contrastive based 3D action representation learning has made great progress. However, the strict positive/negative constraint is yet to be relaxed and the use of non-self positive is yet to be explored. In this paper, a Contrastive Positive Mining (CPM) framework is proposed for unsupervised skeleton 3D action representation learning. The CPM identifies non-self positives in a contextual queue to boost learning. Specifically, the siamese encoders are adopted and trained to match the similarity distributions of the augmented instances in reference to all instances in the contextual queue. By identifying the non-self positive instances in the queue, a positive-enhanced learning strategy is proposed to leverage the knowledge of mined positives to boost the robustness of the learned latent space against intra-class and inter-class diversity. Experimental results have shown that the proposed CPM is effective and outperforms the existing state-of-the-art unsupervised methods on the challenging NTU and PKU-MMD datasets.

Publication Date


  • 2022

Edition


Citation


  • Zhang, H., Hou, Y., Zhang, W., & Li, W. (2022). Contrastive Positive Mining for��Unsupervised 3D Action Representation Learning. In Unknown Book (Vol. 13664 LNCS, pp. 36-51). doi:10.1007/978-3-031-19772-7_3

International Standard Book Number (isbn) 13


  • 9783031197710

Scopus Eid


  • 2-s2.0-85142712568

Web Of Science Accession Number


Book Title


  • Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Start Page


  • 36

End Page


  • 51

Place Of Publication


Abstract


  • Recent contrastive based 3D action representation learning has made great progress. However, the strict positive/negative constraint is yet to be relaxed and the use of non-self positive is yet to be explored. In this paper, a Contrastive Positive Mining (CPM) framework is proposed for unsupervised skeleton 3D action representation learning. The CPM identifies non-self positives in a contextual queue to boost learning. Specifically, the siamese encoders are adopted and trained to match the similarity distributions of the augmented instances in reference to all instances in the contextual queue. By identifying the non-self positive instances in the queue, a positive-enhanced learning strategy is proposed to leverage the knowledge of mined positives to boost the robustness of the learned latent space against intra-class and inter-class diversity. Experimental results have shown that the proposed CPM is effective and outperforms the existing state-of-the-art unsupervised methods on the challenging NTU and PKU-MMD datasets.

Publication Date


  • 2022

Edition


Citation


  • Zhang, H., Hou, Y., Zhang, W., & Li, W. (2022). Contrastive Positive Mining for��Unsupervised 3D Action Representation Learning. In Unknown Book (Vol. 13664 LNCS, pp. 36-51). doi:10.1007/978-3-031-19772-7_3

International Standard Book Number (isbn) 13


  • 9783031197710

Scopus Eid


  • 2-s2.0-85142712568

Web Of Science Accession Number


Book Title


  • Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Start Page


  • 36

End Page


  • 51

Place Of Publication