Efficient and Effective Weakly-Supervised Action Segmentation via Action-Transition-Aware Boundary Alignment

arXiv:2403.19225v1 Announce Type: new
Abstract: Weakly-supervised action segmentation is a task of learning to partition a long video into several action segments, where training videos are only accompanied by transcripts (ordered list of actions). Most of existing methods need to infer pseudo segmentation for training by serial alignment between all frames and the transcript, which is time-consuming and hard to be parallelized while training. In this work, we aim to escape from this inefficient alignment with massive but redundant frames, …

abstract alignment arxiv cs.cv list segmentation training transcripts transition type via video videos weakly-supervised