Learning to Localize Sound Source in Visual Scenes
This addresses the challenge of cross-modal learning for sound source localization in AI, but it is incremental as it builds on existing unsupervised and supervised frameworks.
The paper tackles the problem of localizing sound sources in visual scenes using an unsupervised algorithm, showing that while unsupervised methods can lead to false conclusions, adding minimal supervision corrects these errors and enables effective localization.
Visual events are usually accompanied by sounds in our daily lives. We pose the question: Can the machine learn the correspondence between visual scene and the sound, and localize the sound source only by observing sound and visual scene pairs like human? In this paper, we propose a novel unsupervised algorithm to address the problem of localizing the sound source in visual scenes. A two-stream network structure which handles each modality, with attention mechanism is developed for sound source localization. Moreover, although our network is formulated within the unsupervised learning framework, it can be extended to a unified architecture with a simple modification for the supervised and semi-supervised learning settings as well. Meanwhile, a new sound source dataset is developed for performance evaluation. Our empirical evaluation shows that the unsupervised method eventually go through false conclusion in some cases. We show that even with a few supervision, false conclusion is able to be corrected and the source of sound in a visual scene can be localized effectively.