Self-Supervised Learning by Cross-Modal Audio-Video Clustering
This addresses the need for effective self-supervised learning methods in multi-modal AI, offering a novel approach that surpasses supervised pretraining for action recognition, though it is incremental in leveraging cross-modal correlations.
The paper tackles the problem of self-supervised learning for video and audio representations by proposing Cross-Modal Deep Clustering (XDC), which uses unsupervised clustering in one modality as a supervisory signal for the other, achieving state-of-the-art accuracy on multiple benchmarks and outperforming fully-supervised pretraining on action recognition tasks like HMDB51 and UCF101.
Visual and audio modalities are highly correlated, yet they contain different information. Their strong correlation makes it possible to predict the semantics of one from the other with good accuracy. Their intrinsic differences make cross-modal prediction a potentially more rewarding pretext task for self-supervised learning of video and audio representations compared to within-modality learning. Based on this intuition, we propose Cross-Modal Deep Clustering (XDC), a novel self-supervised method that leverages unsupervised clustering in one modality (e.g., audio) as a supervisory signal for the other modality (e.g., video). This cross-modal supervision helps XDC utilize the semantic correlation and the differences between the two modalities. Our experiments show that XDC outperforms single-modality clustering and other multi-modal variants. XDC achieves state-of-the-art accuracy among self-supervised methods on multiple video and audio benchmarks. Most importantly, our video model pretrained on large-scale unlabeled data significantly outperforms the same model pretrained with full-supervision on ImageNet and Kinetics for action recognition on HMDB51 and UCF101. To the best of our knowledge, XDC is the first self-supervised learning method that outperforms large-scale fully-supervised pretraining for action recognition on the same architecture.