SonoWorld:从单张图像到三维视听场景
SonoWorld: From One Image to a 3D Audio-Visual Scene
把单图三维场景生成从纯视觉扩展到视听联合:先用外扩得到 360 度全景,再提升为可导航三维场景,并在场景几何上放置语言引导的声源锚点、渲染 ambisonics 空间音频,为做 3D 生成的人提供了一个可复用的『几何对齐空间音频』管线与评测设定。
原文依据 we introduce Image2AVScene, the task of generating a 3D audio-visual scene from a single image
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉场景生成领域的巨大进展如今已能把单张图像变成可探索的三维世界,但缺少声音时沉浸感仍不完整。我们提出 Image2AVScene,即从单张图像生成三维视听场景的任务,并给出 SonoWorld——首个应对该挑战的框架。从一张图像出发,我们的流程外扩出 360 度全景图,将其提升为可导航的三维场景,放置语言引导的声音锚点,并为点声源、面声源与环境声源渲染 ambisonics,从而得到与场景几何和语义对齐的空间音频。在新构建的真实世界数据集上的定量评估以及一项受控用户研究,验证了本方法的有效性。除自由视点视听渲染之外,我们还展示了其在单样本声学学习与视听空间声源分离上的应用。项目主页:https://humathe.github.io/sonoworld/
英文原文摘要 914 字符
Tremendous progress in visual scene generation now turns a single image into an explorable 3D world, yet immersion remains incomplete without sound. We introduce Image2AVScene, the task of generating a 3D audio-visual scene from a single image, and present SonoWorld, the first framework to tackle this challenge. From one image, our pipeline outpaints a 360deg panorama, lifts it into a navigable 3D scene, places language-guided sound anchors, and renders ambisonics for point, areal, and ambient sources, yielding spatial audio aligned with scene geometry and semantics. Quantitative evaluations on a newly curated real-world dataset and a controlled user study confirm the effectiveness of our approach. Beyond free-viewpoint audio-visual rendering, we also demonstrate applications to one-shot acoustic learning and audio-visual spatial source separation. Project website: https://humathe.github.io/sonoworld/