SoundBrush: Sound as a Brush for Visual Scene Editing

TL;DR

SoundBrush introduces a novel approach that uses audio cues to guide and enhance visual scene editing within a generative framework, enabling realistic sound-driven modifications and 3D scene extensions.

Contribution

We develop SoundBrush, a model that integrates audio features into a latent diffusion framework for precise visual scene editing guided by sound.

Findings

Successfully manipulates visual scenes based on sound inputs.

Enables insertion of sounding objects matching audio cues.

Extends to sound-driven 3D scene editing.

Abstract

We propose SoundBrush, a model that uses sound as a brush to edit and manipulate visual scenes. We extend the generative capabilities of the Latent Diffusion Model (LDM) to incorporate audio information for editing visual scenes. Inspired by existing image-editing works, we frame this task as a supervised learning problem and leverage various off-the-shelf models to construct a sound-paired visual scene dataset for training. This richly generated dataset enables SoundBrush to learn to map audio features into the textual space of the LDM, allowing for visual scene editing guided by diverse in-the-wild sound. Unlike existing methods, SoundBrush can accurately manipulate the overall scenery or even insert sounding objects to best match the audio inputs while preserving the original content. Furthermore, by integrating with novel view synthesis techniques, our framework can be extended to edit 3D scenes, facilitating sound-driven 3D scene manipulation. Demos are available at https://soundbrush.github.io/.