Action Shuffle Alternating Learning for Unsupervised Action Segmentation
This work addresses the problem of segmenting actions in videos without supervision, which is incremental as it builds on prior feature embedding approaches by incorporating action-level structure.
The paper tackles unsupervised action segmentation by introducing a self-supervised learning method that captures both frame- and action-level temporal structures, achieving superior results on datasets like Breakfast, YouTube Instructions, and 50Salads compared to state-of-the-art methods.
This paper addresses unsupervised action segmentation. Prior work captures the frame-level temporal structure of videos by a feature embedding that encodes time locations of frames in the video. We advance prior work with a new self-supervised learning (SSL) of a feature embedding that accounts for both frame- and action-level structure of videos. Our SSL trains an RNN to recognize positive and negative action sequences, and the RNN's hidden layer is taken as our new action-level feature embedding. The positive and negative sequences consist of action segments sampled from videos, where in the former the sampled action segments respect their time ordering in the video, and in the latter they are shuffled. As supervision of actions is not available and our SSL requires access to action segments, we specify an HMM that explicitly models action lengths, and infer a MAP action segmentation with the Viterbi algorithm. The resulting action segmentation is used as pseudo-ground truth for estimating our action-level feature embedding and updating the HMM. We alternate the above steps within the Generalized EM framework, which ensures convergence. Our evaluation on the Breakfast, YouTube Instructions, and 50Salads datasets gives superior results to those of the state of the art.