English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

In 3D action recognition, there exists rich complementary information betweenskeleton modalities. Nevertheless, how to model and utilize this informationremains a challenging problem for self-supervised 3D action representationlearning. In this work, we formulate the cross-modal interaction as abidirectional knowledge distillation problem. Different from classicdistillation solutions that transfer the knowledge of a fixed and pre-trainedteacher to the student, in this work, the knowledge is continuously updated andbidirectionally distilled between modalities. To this end, we propose a newCross-modal Mutual Distillation (CMD) framework with the following designs. Onthe one hand, the neighboring similarity distribution is introduced to modelthe knowledge learned in each modality, where the relational information isnaturally suitable for the contrastive frameworks. On the other hand,asymmetrical configurations are used for teacher and student to stabilize thedistillation process and to transfer high-confidence information betweenmodalities. By derivation, we find that the cross-modal positive mining inprevious works can be regarded as a degenerated version of our CMD. We performextensive experiments on NTU RGB+D 60, NTU RGB+D 120, and PKU-MMD II datasets.Our approach outperforms existing self-supervised methods and sets a series ofnew records. The code is available at: https://github.com/maoyunyao/CMD

CMD: Self-supervised 3D Action Representation Learning with Cross-modal Mutual Distillation