Structure from Silence: Learning Scene Structure from Ambient Sound

Chen, Ziyang; Hu, Xuedong; Owens, Andrew

doi:10.48550/arxiv.2111.05846

Cited by 1 publication

(1 citation statement)

References 67 publications

(86 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Purushwalkam et al [69] reconstructed floor plans in simulated environments [12]. Chen et al [18] used ambient sounds from environments to learn about scene structures. Konno et al [46] integrated sound localization to visual SfM while do not jointly learn them.…”

Section: Related Workmentioning

confidence: 99%

Sound Localization from Motion: Jointly Learning Sound Direction and Camera Rotation

Chen¹,

Qian²,

Owens³

2023

Preprint

View full text Add to dashboard Cite

The images and sounds that we perceive undergo subtle but geometrically consistent changes as we rotate our heads. In this paper, we use these cues to solve a problem we call Sound Localization from Motion (SLfM): jointly estimating camera rotation and localizing sound sources. We learn to solve these tasks solely through self-supervision. A visual model predicts camera rotation from a pair of images, while an audio model predicts the direction of sound sources from binaural sounds. We train these models to generate predictions that agree with one another. At test time, the models can be deployed independently. To obtain a feature representation that is well-suited to solving this challenging problem, we also propose a method for learning an audio-visual representation through cross-view binauralization: estimating binaural sound from one view, given images and sound from another. Our model can successfully estimate accurate rotations on both real and synthetic scenes, and localize sound sources with accuracy competitive with state-of-the-art self-supervised approaches. Project site: https://ificl.github.io/SLfM .

show abstract