World-Shaper: A Unified Framework for 360{\deg} Panoramic Editing

2026-01-30

AI Summary

Existing panoramic image editing methods struggle to effectively model spatial structure while preserving global spherical geometric consistency. This work proposes the first unified generation and editing framework operating directly in the equirectangular projection (ERP) domain, introducing a geometry-aware learning strategy that combines explicit position-aware shape supervision with implicit internalization of panoramic priors. To address the scarcity of paired training data, the method adopts a generate-then-edit paradigm and employs progressive training to enhance stability. Evaluated on the newly introduced PEBench benchmark, the proposed approach significantly outperforms existing methods in geometric consistency, editing fidelity, and text controllability, enabling high-quality, coherently controlled 360° visual content generation.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation
• Machine Learning: Learning with Manifolds

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Being able to edit panoramic images is crucial for creating realistic 360{\deg} visual experiences. However, existing perspective-based image editing methods fail to model the spatial structure of panoramas. Conventional cube-map decompositions attempt to overcome this problem but inevitably break global consistency due to their mismatch with spherical geometry. Motivated by this insight, we reformulate panoramic editing directly in the equirectangular projection (ERP) domain and present World-Shaper, a unified geometry-aware framework that bridges panoramic generation and editing within a single editing-centric design. To overcome the scarcity of paired data, we adopt a generate-then-edit paradigm, where controllable panoramic generation serves as an auxiliary stage to synthesize diverse paired examples for supervised editing learning. To address geometric distortion, we introduce a geometry-aware learning strategy that explicitly enforces position-aware shape supervision and implicitly internalizes panoramic priors through progressive training. Extensive experiments on our new benchmark, PEBench, demonstrate that our method achieves superior geometric consistency, editing fidelity, and text controllability compared to SOTA methods, enabling coherent and flexible 360{\deg} visual world creation with unified editing control. Code, model, and data will be released at our project page: https://world-shaper-project.github.io/

Problem

• panoramic editing
• geometric consistency
• equirectangular projection
• 360° visual experience
• spherical geometry

Innovation

• panoramic editing
• equirectangular projection
• geometry-aware learning
• generate-then-edit
• 360° visual consistency