PanoDreamer: Optimization-Based Single Image to 360 3D Scene With Diffusion
This addresses the challenge of 3D scene reconstruction for applications like virtual reality or gaming, but appears incremental as it builds on existing single-image methods with optimization-based improvements.
The paper tackles the problem of generating a coherent 360° 3D scene from a single input image by framing it as single-image panorama and depth estimation, and demonstrates that their approach outperforms existing techniques in consistency and overall quality.
In this paper, we present PanoDreamer, a novel method for producing a coherent 360° 3D scene from a single input image. Unlike existing methods that generate the scene sequentially, we frame the problem as single-image panorama and depth estimation. Once the coherent panoramic image and its corresponding depth are obtained, the scene can be reconstructed by inpainting the small occluded regions and projecting them into 3D space. Our key contribution is formulating single-image panorama and depth estimation as two optimization tasks and introducing alternating minimization strategies to effectively solve their objectives. We demonstrate that our approach outperforms existing techniques in single-image 360° 3D scene reconstruction in terms of consistency and overall quality.