360Anything: Geometry-Free Lifting of Images and Videos to 360{\deg}

TL;DR

360Anything is a data-driven, geometry-free framework that uses pre-trained diffusion transformers to lift images and videos to 360-degree panoramas without needing camera metadata, achieving state-of-the-art results.

Contribution

It introduces a novel geometry-free approach using diffusion transformers for perspective-to-360 panorama conversion, removing the dependence on camera calibration data.

Findings

State-of-the-art performance on image and video perspective-to-360 generation

Outperforms methods requiring ground-truth camera information

Demonstrates strong zero-shot camera FoV and orientation estimation

Abstract

Lifting perspective images and videos to 360{\deg} panoramas enables immersive 3D world generation. Existing approaches often rely on explicit geometric alignment between the perspective and the equirectangular projection (ERP) space. Yet, this requires known camera metadata, obscuring the application to in-the-wild data where such calibration is typically absent or noisy. We propose 360Anything, a geometry-free framework built upon pre-trained diffusion transformers. By treating the perspective input and the panorama target simply as token sequences, 360Anything learns the perspective-to-equirectangular mapping in a purely data-driven way, eliminating the need for camera information. Our approach achieves state-of-the-art performance on both image and video perspective-to-360{\deg} generation, outperforming prior works that use ground-truth camera information. We also trace the root cause of the seam artifacts at ERP boundaries to zero-padding in the VAE encoder, and introduce Circular Latent Encoding to facilitate seamless generation. Finally, we show competitive results in zero-shot camera FoV and orientation estimation benchmarks, demonstrating 360Anything's deep geometric understanding and broader utility in computer vision tasks. Additional results are available at https://360anything.github.io/.