Syntactically Guided Generative Embeddings for Zero-Shot Skeleton Action Recognition
This work addresses the problem of recognizing skeleton-based actions without labeled examples for new classes, which is incremental as it builds on existing zero-shot learning methods but applies them to a specific domain with novel constraints.
The paper tackles zero-shot skeleton action recognition by introducing SynSE, a syntactically guided generative approach that learns refined embedding spaces constrained across visual and language modalities, achieving state-of-the-art performance on NTU-60 and NTU-120 datasets in both ZSL and GZSL settings.
We introduce SynSE, a novel syntactically guided generative approach for Zero-Shot Learning (ZSL). Our end-to-end approach learns progressively refined generative embedding spaces constrained within and across the involved modalities (visual, language). The inter-modal constraints are defined between action sequence embedding and embeddings of Parts of Speech (PoS) tagged words in the corresponding action description. We deploy SynSE for the task of skeleton-based action sequence recognition. Our design choices enable SynSE to generalize compositionally, i.e., recognize sequences whose action descriptions contain words not encountered during training. We also extend our approach to the more challenging Generalized Zero-Shot Learning (GZSL) problem via a confidence-based gating mechanism. We are the first to present zero-shot skeleton action recognition results on the large-scale NTU-60 and NTU-120 skeleton action datasets with multiple splits. Our results demonstrate SynSE's state of the art performance in both ZSL and GZSL settings compared to strong baselines on the NTU-60 and NTU-120 datasets. The code and pretrained models are available at https://github.com/skelemoa/synse-zsl