Joint Visual-Temporal Embedding for Unsupervised Learning of Actions in Untrimmed Sequences

Rosaura G. VidalMata, Walter J. Scheirer, Anna Kukleva, David Cox, Hilde Kuehne; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2021, pp. 1238-1247

Abstract


Understanding the structure of complex activities in untrimmed videos is a challenging task in the area of action recognition. One problem here is that this task usually requires a large amount of hand-annotated minute- or even hour-long video data, but annotating such data is very time consuming and can not easily be automated or scaled. To address this problem, this paper proposes an approach for the unsupervised learning of actions in untrimmed video sequences based on a joint visual-temporal embedding space. To this end, we combine a visual embedding based on a predictive U-Net architecture with a temporal continuous function. The resulting representation space allows detecting relevant action clusters based on their visual as well as their temporal appearance. The proposed method is evaluated on three standard benchmark datasets, Breakfast Actions, INRIA YouTube Instructional Videos, and 50 Salads. We show that the proposed approach is able to provide a meaningful visual and temporal embedding out of the visual cues present in contiguous video frames and is suitable for the task of unsupervised temporal segmentation of actions.

Related Material


[pdf] [arXiv]
[bibtex]
@InProceedings{VidalMata_2021_WACV, author = {VidalMata, Rosaura G. and Scheirer, Walter J. and Kukleva, Anna and Cox, David and Kuehne, Hilde}, title = {Joint Visual-Temporal Embedding for Unsupervised Learning of Actions in Untrimmed Sequences}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2021}, pages = {1238-1247} }