arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2301.08730cs.CVcs.SDeess.AS
收藏

Novel-View Acoustic Synthesis(新视角声学合成)

Novel-View Acoustic Synthesis

University of Texas at Austin(德克萨斯大学奥斯汀分校);FAIR, Meta AI(Meta AI 基础人工智能研究团队);Reality Labs Research at Meta(Meta 现实实验室研究部)

Changan Chen, Alexander Richard, Roman Shapovalov, Vamsi Krishna Ithapu, Natalia Neverova, Kristen Grauman, Andrea Vedaldi

AI总结:

提出新视角声学合成任务及ViGAS网络,通过分析视听线索合成目标视点声音,并构建大规模数据集验证有效性,推动多模态新视角合成研究。

摘要:

我们引入了novel-view acoustic synthesis(NVAS,新视角声学合成)任务:给定在源视点观察到的视觉和声音,能否从未见过的目标视点合成该场景的声音?我们提出了一种神经渲染方法:Visually-Guided Acoustic Synthesis(ViGAS,视觉引导声学合成)网络,它通过分析输入的视听线索来学习合成空间中任意点的声音。为了对该任务进行基准测试,我们收集了两个首创的大规模多视角视听数据集,一个为合成数据集,另一个为真实数据集。我们表明,我们的模型能够成功推理空间线索,并在两个数据集上合成忠实的音频。据我们所知,这项工作是解决novel-view acoustic synthesis任务的首次系统化表述、数据集构建和方法提出,具有从AR/VR到艺术与设计等激动人心的潜在应用。借助本研究的突破,我们相信novel-view synthesis(新视角合成)的未来在于从视频中进行多模态学习。

英文摘要:

We introduce the novel-view acoustic synthesis (NVAS) task: given the sight and sound observed at a source viewpoint, can we synthesize the sound of that scene from an unseen target viewpoint? We propose a neural rendering approach: Visually-Guided Acoustic Synthesis (ViGAS) network that learns to synthesize the sound of an arbitrary point in space by analyzing the input audio-visual cues. To benchmark this task, we collect two first-of-their-kind large-scale multi-view audio-visual datasets, one synthetic and one real. We show that our model successfully reasons about the spatial cues and synthesizes faithful audio on both datasets. To our knowledge, this work represents the very first formulation, dataset, and approach to solve the novel-view acoustic synthesis task, which has exciting potential applications ranging from AR/VR to art and design. Unlocked by this work, we believe that the future of novel-view synthesis is in multi-modal learning from videos.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑