Hearing Hands: Generating Sounds from Physical Interactions in 3D Scenes

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dou, Yiming, Oh, Wonseok, Luo, Yuqing, Loquercio, Antonio, Owens, Andrew
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913889435779072
author Dou, Yiming
Oh, Wonseok
Luo, Yuqing
Loquercio, Antonio
Owens, Andrew
author_facet Dou, Yiming
Oh, Wonseok
Luo, Yuqing
Loquercio, Antonio
Owens, Andrew
contents We study the problem of making 3D scene reconstructions interactive by asking the following question: can we predict the sounds of human hands physically interacting with a scene? First, we record a video of a human manipulating objects within a 3D scene using their hands. We then use these action-sound pairs to train a rectified flow model to map 3D hand trajectories to their corresponding audio. At test time, a user can query the model for other actions, parameterized as sequences of hand poses, to estimate their corresponding sounds. In our experiments, we find that our generated sounds accurately convey material properties and actions, and that they are often indistinguishable to human observers from real sounds. Project page: https://www.yimingdou.com/hearing_hands/
format Preprint
id arxiv_https___arxiv_org_abs_2506_09989
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hearing Hands: Generating Sounds from Physical Interactions in 3D Scenes
Dou, Yiming
Oh, Wonseok
Luo, Yuqing
Loquercio, Antonio
Owens, Andrew
Computer Vision and Pattern Recognition
We study the problem of making 3D scene reconstructions interactive by asking the following question: can we predict the sounds of human hands physically interacting with a scene? First, we record a video of a human manipulating objects within a 3D scene using their hands. We then use these action-sound pairs to train a rectified flow model to map 3D hand trajectories to their corresponding audio. At test time, a user can query the model for other actions, parameterized as sequences of hand poses, to estimate their corresponding sounds. In our experiments, we find that our generated sounds accurately convey material properties and actions, and that they are often indistinguishable to human observers from real sounds. Project page: https://www.yimingdou.com/hearing_hands/
title Hearing Hands: Generating Sounds from Physical Interactions in 3D Scenes
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.09989