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

Avinash Paliwal, Xilong Zhou, Andrii Tsarov, Nima Khademi Kalantari
12/6/2024
cs.CVcs.GR

No official code found. Best implementation: avinashpaliwal/PanoDreamer (109 stars, last commit 2026-04-24).

Abstract

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.

View on arXivView PDF

Code Implementations(1)

avinashpaliwal/PanoDreamerCommunity100%

[SIGGRAPH Asia 2025] PanoDreamer: Optimization-Based Single Image to 360 3D Scene With Diffusion

1098PythonDec 6, 20246 months ago
3d-from-a-single-image3d-scene-generationdiffusion-modelgaussian-splattingpanorama-depth+3 more

Similar Papers

AuraFusion360: Augmented Unseen Region Alignment for Reference-based 360° Unbounded Scene Inpainting

86% match2026

Generating 360° Video is What You Need For a 3D Scene

85% match2025

Robust and Flexible Omnidirectional Depth Estimation with Multiple 360-degree Cameras

84% match2024

VGGT-360: Geometry-Consistent Zero-Shot Panoramic Depth Estimation

84% match2026

Inpaint360GS: Efficient Object-Aware 3D Inpainting via Gaussian Splatting for 360° Scenes

84% match1 repo492025

360Recon: An Accurate Reconstruction Method Based on Depth Fusion from 360 Images

84% match1 repo192024

Cite this paper

@article{paliwal2024panodreamer,
  title  = {PanoDreamer: Optimization-Based Single Image to 360 3D Scene With Diffusion},
  author = {Avinash Paliwal and Xilong Zhou and Andrii Tsarov and Nima Khademi Kalantari},
  year   = {2024},
  eprint = {2412.04827},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2412.04827v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2412.04827.svg)](https://codeofpaper.com/paper/2412.04827)

Discussion