Time-Equivariant Contrastive Video Representation Learning
This addresses the challenge of capturing temporal information in videos for tasks like retrieval and recognition, offering a novel approach that improves over existing methods.
The paper tackles the problem of learning video representations from unlabeled data by proposing a self-supervised contrastive method that preserves video dynamics and is equivariant to temporal transformations, achieving state-of-the-art results in video retrieval and action recognition on benchmarks like UCF101, HMDB51, and Diving48.
We introduce a novel self-supervised contrastive learning method to learn representations from unlabelled videos. Existing approaches ignore the specifics of input distortions, e.g., by learning invariance to temporal transformations. Instead, we argue that video representation should preserve video dynamics and reflect temporal manipulations of the input. Therefore, we exploit novel constraints to build representations that are equivariant to temporal transformations and better capture video dynamics. In our method, relative temporal transformations between augmented clips of a video are encoded in a vector and contrasted with other transformation vectors. To support temporal equivariance learning, we additionally propose the self-supervised classification of two clips of a video into 1. overlapping 2. ordered, or 3. unordered. Our experiments show that time-equivariant representations achieve state-of-the-art results in video retrieval and action recognition benchmarks on UCF101, HMDB51, and Diving48.