Diffusion-Based Attention Warping for Consistent 3D Scene Editing
This addresses the challenge of coherent 3D scene manipulation for applications like virtual reality or content creation, representing a strong specific gain in this domain.
The paper tackles the problem of ensuring view consistency and realism in 3D scene editing by using diffusion models with attention features warped across views based on geometry, achieving high fidelity and spatial alignment.
We present a novel method for 3D scene editing using diffusion models, designed to ensure view consistency and realism across perspectives. Our approach leverages attention features extracted from a single reference image to define the intended edits. These features are warped across multiple views by aligning them with scene geometry derived from Gaussian splatting depth estimates. Injecting these warped features into other viewpoints enables coherent propagation of edits, achieving high fidelity and spatial alignment in 3D space. Extensive evaluations demonstrate the effectiveness of our method in generating versatile edits of 3D scenes, significantly advancing the capabilities of scene manipulation compared to the existing methods. Project page: \url{https://attention-warp.github.io}