Discriminative Sounding Objects Localization via Self-supervised Audiovisual Matching
This work addresses the challenge of discriminative sounding object localization in cocktail-party scenarios, which is incremental by building on self-supervised audiovisual matching techniques.
The paper tackles the problem of localizing sounding objects in mixed sound scenes (cocktail-party) by proposing a two-stage self-supervised learning framework that aggregates object representations from single-source scenes and matches audio-visual category distributions, achieving superior performance in filtering out silent objects and locating sounding ones in both realistic and synthesized videos.
Discriminatively localizing sounding objects in cocktail-party, i.e., mixed sound scenes, is commonplace for humans, but still challenging for machines. In this paper, we propose a two-stage learning framework to perform self-supervised class-aware sounding object localization. First, we propose to learn robust object representations by aggregating the candidate sound localization results in the single source scenes. Then, class-aware object localization maps are generated in the cocktail-party scenarios by referring the pre-learned object knowledge, and the sounding objects are accordingly selected by matching audio and visual object category distributions, where the audiovisual consistency is viewed as the self-supervised signal. Experimental results in both realistic and synthesized cocktail-party videos demonstrate that our model is superior in filtering out silent objects and pointing out the location of sounding objects of different classes. Code is available at https://github.com/DTaoo/Discriminative-Sounding-Objects-Localization.