BID: Boundary-Interior Decoding for Unsupervised Temporal Action Localization Pre-Trainin
This addresses the need for robust action localization in video analysis, particularly for skeleton-based data, though it appears incremental as it builds on existing pre-training paradigms.
The paper tackles the problem of ambiguous and incomplete skeleton-based motion representations for temporal action localization by proposing BID, the first unsupervised pre-training framework that partitions motion sequences into semantically meaningful pre-action segments. When fine-tuned with minimal annotated data, BID outperforms state-of-the-art methods by a large margin.
Skeleton-based motion representations are robust for action localization and understanding for their invariance to perspective, lighting, and occlusion, compared with images. Yet, they are often ambiguous and incomplete when taken out of context, even for human annotators. As infants discern gestures before associating them with words, actions can be conceptualized before being grounded with labels. Therefore, we propose the first unsupervised pre-training framework, Boundary-Interior Decoding (BID), that partitions a skeleton-based motion sequence into discovered semantically meaningful pre-action segments. By fine-tuning our pre-training network with a small number of annotated data, we show results out-performing SOTA methods by a large margin.