Self-supervised Co-training for Video Representation Learning
This addresses the problem of data efficiency in video representation learning for computer vision researchers, though it is incremental as it builds on existing contrastive learning methods.
The paper tackles self-supervised video representation learning by proposing a co-training scheme that uses complementary views (RGB and optical flow) to improve InfoNCE loss, achieving state-of-the-art or comparable performance on action recognition and video retrieval while being more efficient, requiring far less training data.
The objective of this paper is visual-only self-supervised video representation learning. We make the following contributions: (i) we investigate the benefit of adding semantic-class positives to instance-based Info Noise Contrastive Estimation (InfoNCE) training, showing that this form of supervised contrastive learning leads to a clear improvement in performance; (ii) we propose a novel self-supervised co-training scheme to improve the popular infoNCE loss, exploiting the complementary information from different views, RGB streams and optical flow, of the same data source by using one view to obtain positive class samples for the other; (iii) we thoroughly evaluate the quality of the learnt representation on two different downstream tasks: action recognition and video retrieval. In both cases, the proposed approach demonstrates state-of-the-art or comparable performance with other self-supervised approaches, whilst being significantly more efficient to train, i.e. requiring far less training data to achieve similar performance.