マルチモーダル物理シーン理解のための分離音響場
Disentangled Acoustic Fields For Multimodal Physical Scene Understanding
音の生成・伝播過程を分離表現でモデル化し、落下物の位置に関する空間不確実性マップを構築して探索候補を提示することで、物体探索の成功率を高める手法を提案。
著者: Jie Yin, Andrew Luo, Yilun Du, Anoop Cherian, Tim K. Marks, Jonathan Le Roux, Chuang Gan
分類: cs.RO, cs.SD, eess.AS
原文アブストラクト
We study the problem of multimodal physical scene understanding, where an embodied agent needs to find fallen objects by inferring object properties, direction, and distance of an impact sound source. Previous works adopt feed-forward neural networks to directly regress the variables from sound, leading to poor generalization and domain adaptation issues. In this paper, we illustrate that learning a disentangled model of acoustic formation, referred to as disentangled acoustic field (DAF), to capture the sound generation and propagation process, enables the embodied agent to construct a spatial uncertainty map over where the objects may have fallen. We demonstrate that our analysis-by-synthesis framework can jointly infer sound properties by explicitly decomposing and factorizing the latent space of the disentangled model. We further show that the spatial uncertainty map can significantly improve the success rate for the localization of fallen objects by proposing multiple plausible exploration locations.
