PanoDreamer: Optimization-Based Single Image to 360 3D Scene With Diffusion

TL;DR

PanoDreamer is a novel optimization-based method that reconstructs a coherent 360-degree 3D scene from a single image by estimating panorama and depth through alternating minimization, outperforming existing techniques.

Contribution

It introduces a new formulation of panorama and depth estimation as optimization tasks with alternating minimization, enabling better scene coherence from a single image.

Findings

Outperforms existing methods in scene consistency

Produces higher quality 360-degree reconstructions

Effective in handling occlusions and inpainting

Abstract

In this paper, we present PanoDreamer, a novel method for producing a coherent 360{\deg} 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{\deg} 3D scene reconstruction in terms of consistency and overall quality.