Unsupervised Action Segmentation by Joint Representation Learning and Online Clustering
This addresses the problem of segmenting activities in videos without labeled data for researchers in computer vision, offering an incremental improvement by combining representation learning and clustering in an online manner.
The paper tackles unsupervised activity segmentation in videos by jointly learning representations and performing online clustering, using temporal optimal transport to preserve activity order. It achieves performance on par with or better than previous methods on datasets like 50-Salads, YouTube Instructions, Breakfast, and Desktop Assembly, with significantly reduced memory constraints.
We present a novel approach for unsupervised activity segmentation which uses video frame clustering as a pretext task and simultaneously performs representation learning and online clustering. This is in contrast with prior works where representation learning and clustering are often performed sequentially. We leverage temporal information in videos by employing temporal optimal transport. In particular, we incorporate a temporal regularization term which preserves the temporal order of the activity into the standard optimal transport module for computing pseudo-label cluster assignments. The temporal optimal transport module enables our approach to learn effective representations for unsupervised activity segmentation. Furthermore, previous methods require storing learned features for the entire dataset before clustering them in an offline manner, whereas our approach processes one mini-batch at a time in an online manner. Extensive evaluations on three public datasets, i.e. 50-Salads, YouTube Instructions, and Breakfast, and our dataset, i.e., Desktop Assembly, show that our approach performs on par with or better than previous methods, despite having significantly less memory constraints. Our code and dataset are available on our research website: https://retrocausal.ai/research/