Separating the "Chirp" from the "Chat": Self-supervised Visual Grounding of Sound and Language
This work addresses the challenge of automatically discovering and distinguishing audio-visual associations in videos, which is important for applications in multimedia analysis and AI perception, representing a novel method rather than an incremental improvement.
The paper tackles the problem of self-supervised visual grounding of sound and language in videos, introducing DenseAV, which learns to localize words and sounds without explicit supervision and outperforms prior methods like ImageBind on cross-modal retrieval with fewer parameters.
We present DenseAV, a novel dual encoder grounding architecture that learns high-resolution, semantically meaningful, and audio-visually aligned features solely through watching videos. We show that DenseAV can discover the ``meaning'' of words and the ``location'' of sounds without explicit localization supervision. Furthermore, it automatically discovers and distinguishes between these two types of associations without supervision. We show that DenseAV's localization abilities arise from a new multi-head feature aggregation operator that directly compares dense image and audio representations for contrastive learning. In contrast, many other systems that learn ``global'' audio and video representations cannot localize words and sound. Finally, we contribute two new datasets to improve the evaluation of AV representations through speech and sound prompted semantic segmentation. On these and other datasets we show DenseAV dramatically outperforms the prior art on speech and sound prompted semantic segmentation. DenseAV outperforms the previous state-of-the-art, ImageBind, on cross-modal retrieval using fewer than half of the parameters. Project Page: \href{https://aka.ms/denseav}{https://aka.ms/denseav}