Look, Listen and Learn
It addresses the problem of unsupervised representation learning for researchers in computer vision and audio processing, offering a novel approach but with incremental gains in specific domains.
The paper tackled learning from unlabeled videos by leveraging audio-visual correspondence, resulting in state-of-the-art performance on sound classification benchmarks and competitive results on ImageNet classification.
We consider the question: what can be learnt by looking at and listening to a large number of unlabelled videos? There is a valuable, but so far untapped, source of information contained in the video itself -- the correspondence between the visual and the audio streams, and we introduce a novel "Audio-Visual Correspondence" learning task that makes use of this. Training visual and audio networks from scratch, without any additional supervision other than the raw unconstrained videos themselves, is shown to successfully solve this task, and, more interestingly, result in good visual and audio representations. These features set the new state-of-the-art on two sound classification benchmarks, and perform on par with the state-of-the-art self-supervised approaches on ImageNet classification. We also demonstrate that the network is able to localize objects in both modalities, as well as perform fine-grained recognition tasks.