Video Contrastive Learning with Global Context
This work addresses the limitation of existing video contrastive learning methods that rely on short-range spatiotemporal salience, offering a more robust approach for video representation learning.
The paper tackled the problem of video contrastive learning by proposing a video-level method that captures global context and incorporates temporal order regularization, resulting in state-of-the-art performance on five video datasets for action classification, localization, and retrieval.
Contrastive learning has revolutionized self-supervised image representation learning field, and recently been adapted to video domain. One of the greatest advantages of contrastive learning is that it allows us to flexibly define powerful loss objectives as long as we can find a reasonable way to formulate positive and negative samples to contrast. However, existing approaches rely heavily on the short-range spatiotemporal salience to form clip-level contrastive signals, thus limit themselves from using global context. In this paper, we propose a new video-level contrastive learning method based on segments to formulate positive pairs. Our formulation is able to capture global context in a video, thus robust to temporal content change. We also incorporate a temporal order regularization term to enforce the inherent sequential structure of videos. Extensive experiments show that our video-level contrastive learning framework (VCLR) is able to outperform previous state-of-the-arts on five video datasets for downstream action classification, action localization and video retrieval. Code is available at https://github.com/amazon-research/video-contrastive-learning.