AuraFusion360: Augmented Unseen Region Alignment for Reference-based 360° Unbounded Scene Inpainting
This addresses the challenge of high-quality object removal and hole filling in 3D scenes for applications like virtual reality and architectural visualization, representing a novel method for a known bottleneck.
The paper tackles the problem of 3D scene inpainting in 360° unbounded scenes, where existing methods struggle with view consistency and geometric accuracy, and presents AuraFusion360, a reference-based method that achieves superior perceptual quality and maintains geometric accuracy across viewpoint changes, as demonstrated by outperforming existing methods in extensive experiments.
Three-dimensional scene inpainting is crucial for applications from virtual reality to architectural visualization, yet existing methods struggle with view consistency and geometric accuracy in 360° unbounded scenes. We present AuraFusion360, a novel reference-based method that enables high-quality object removal and hole filling in 3D scenes represented by Gaussian Splatting. Our approach introduces (1) depth-aware unseen mask generation for accurate occlusion identification, (2) Adaptive Guided Depth Diffusion, a zero-shot method for accurate initial point placement without requiring additional training, and (3) SDEdit-based detail enhancement for multi-view coherence. We also introduce 360-USID, the first comprehensive dataset for 360° unbounded scene inpainting with ground truth. Extensive experiments demonstrate that AuraFusion360 significantly outperforms existing methods, achieving superior perceptual quality while maintaining geometric accuracy across dramatic viewpoint changes.