Cross-Modal Label Contrastive Learning for Unsupervised Audio-Visual Event Localization

This paper for the first time explores audio-visual event localization in an unsupervised manner. Previous methods tackle this problem in a supervised setting and require segment-level or video-level event category ground-truth to train the model. However, building large-scale multi-modality datasets with category annotations is human-intensive and thus not scalable to real-world applications. To this end, we propose crossmodal label contrastive learning to exploit multi-modal information among unlabeled audio and visual streams as selfsupervision signals. At the feature representation level, multimodal representations are collaboratively learned from audio and visual components by using self-supervised representation learning. At the label level, we propose a novel selfsupervised pretext task i.e. label contrasting to self-annotate videos with pseudo-labels for localization model training. Note that irrelevant background would hinder the acquisition of high-quality pseudo-labels and thus lead to an inferior localization model. To address this issue, we then propose an expectation-maximization algorithm that optimizes the pseudo-label acquisition and localization model in a coarseto-fine manner. Extensive experiments demonstrate that our unsupervised approach performs reasonably well compared to the state-of-the-art supervised methods.