Unsupervised Learning of Video Representations via Dense Trajectory Clustering - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year :

Unsupervised Learning of Video Representations via Dense Trajectory Clustering


This paper addresses the task of unsupervised learning of representations for action recognition in videos. Previous works proposed to utilize future prediction, or other domain-specific objectives to train a network, but achieved only limited success. In contrast, in the relevant field of image representation learning, simpler, discrimination-based methods have recently bridged the gap to fully-supervised performance. We first propose to adapt two top performing objectives in this class-instance recognition and local aggregation, to the video domain. In particular, the latter approach iterates between clustering the videos in the feature space of a network and updating it to respect the cluster with a non-parametric classification loss. We observe promising performance, but qualitative analysis shows that the learned representations fail to capture motion patterns, grouping the videos based on appearance. To mitigate this issue, we turn to the heuristic-based IDT descriptors, that were manually designed to encode motion patterns in videos. We form the clusters in the IDT space, using these descriptors as a an unsupervised prior in the iterative local aggregation algorithm. Our experiments demonstrates that this approach outperform prior work on UCF101 and HMDB51 action recognition benchmarks 3. We also qualitatively analyze the learned representations and show that they successfully capture video dynamics.
Fichier principal
Vignette du fichier
2006.15731.pdf (6.13 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03990585 , version 1 (15-02-2023)



Pavel Tokmakov, Martial Hebert, Cordelia Schmid. Unsupervised Learning of Video Representations via Dense Trajectory Clustering. ECCV Workshop - Visual Inductive Priors for Data-Efficient Deep Learning, Aug 2020, Edinburgh (online), United Kingdom. ⟨hal-03990585⟩
5 View
1 Download



Gmail Facebook Twitter LinkedIn More