TL;DR: Gimbal360 robustly canonicalizes unposed perspective images using differentiable auto-leveling, enabling structurally consistent and seamless 360° panorama generation.
Input Perspective
360° Panorama Generation (ERP)
Seam Visualization
Demo
Input Image
Seamless 360° Panorama Generation
Interactive Viewer
Select an Input
Click an image below to generate its panorama
Abstract
Diffusion models excel at 2D outpainting, but extending them to \(360^\circ\) panoramic completion from unposed perspective images is challenging due to the geometric and topological mismatch between perspective projections and spherical panoramas. We present Gimbal360, a principled framework that explicitly bridges perspective observations and spherical panoramas. We introduce a Canonical Viewing Space that regularizes projective geometry and provides a consistent intermediate representation between the two domains. To anchor in-the-wild inputs to this space, we propose a Differentiable Auto-Leveling module that stabilizes feature orientation without requiring camera parameters at inference. Panoramic generation also introduces a topological challenge. Standard generative architectures assume a bounded Euclidean image plane, while Equirectangular Projection (ERP) panoramas exhibit intrinsic \(S^1\) periodicity. Euclidean operations therefore break boundary continuity. We address this mismatch by enforcing topological equivariance in the latent space to preserve seamless periodic structure. To support this formulation, we introduce Horizon360, a curated large-scale dataset of gravity-aligned panoramic environments. Extensive experiments show that explicitly standardizing geometric and topological priors enables Gimbal360 to achieve state-of-the-art performance in structurally consistent \(360^\circ\) scene completion.
Methodology
Given a perspective sample from our Horizon360 Dataset, the Differentiable Auto-Leveling module predicts a rigid correspondence field to warp the perspective image into a gravity-aligned, yaw-centered Canonical Viewing Space. During Topologically Equivariant Generation, a Siamese Consistency Loss between the standard and horizontally shifted latent streams forces the network to natively respect the continuous \(S^1\) boundary.
BibTeX
@misc{orange2026gimbal360,
title={Gimbal360: Differentiable Auto-Leveling for Canonicalized 360 Panoramic Image Completion},
author={Orange Team},
year={2026},
url={https://orange-3dv-team.github.io/MoCam}
}