Test-Time Zero-Shot Temporal Action Localization

arXiv:2404.05426v1 Announce Type: new
Abstract: Zero-Shot Temporal Action Localization (ZS-TAL) seeks to identify and locate actions in untrimmed videos unseen during training. Existing ZS-TAL methods involve fine-tuning a model on a large amount of annotated training data. While effective, training-based ZS-TAL approaches assume the availability of labeled data for supervised learning, which can be impractical in some applications. Furthermore, the training process naturally induces a domain bias into the learned model, which may adversely affect the model's generalization ability to …

abstract arxiv availability cs.cv data fine-tuning identify localization supervised learning temporal test training training data type videos zero-shot