Mover360: Controllable Object Manipulation in 360° Panoramic Images
2026-08-24 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created Mover360, a tool that helps move, insert, or remove objects in 360° panoramic images, which are tricky because they wrap around and distort differently than normal photos. Their system uses special maps and prompts to let users easily control object placement with just clicks. They trained the model using a custom data pipeline that simulates real and synthetic scenes with accurate lighting and depth. Tests show Mover360 works better than other editing tools for 360° images in keeping the scene realistic and consistent. The authors also provide their code and dataset for others to use.
360° imagesequirectangular projectionobject manipulationdiffusion transformerpanoramic contextimage editingdepth conditioninpaintingsynthetic dataillumination
Authors
Haoyi Zhong, Fang-Lue Zhang, Andrew Chalmers, Taehyun Rhee
Abstract
We present Mover360, a controllable object manipulation framework for 360° images. Unlike perspective images, 360° images in equirectangular projection (ERP) exhibit horizontal wrap-around, latitude-dependent distortion, and global scene continuity, which makes object-level edits difficult for existing perspective editors to produce and for users to specify. To address this, Mover360 centers on object Translation (relocating a specified object within an existing panorama) while supporting reference-guided Insert and Remove as auxiliary tasks. Its interface unifies point-, bbox-, and mask-guided control by encoding each task into a fixed prompt and a compact, ERP-aligned instruction map. In the default point mode, a single click relocates an object, allowing the model to infer a plausible size, support, and illumination using panoramic context and an auxiliary depth condition. Structurally, Mover360 is a lightweight adaptation of a pretrained diffusion transformer. To generate paired supervision, we construct a UE5 data-generation pipeline with surface-aware object placement and randomized illumination, yielding large-scale paired data and a dual-domain benchmark of synthetic and real panoramas with ground truth for all three tasks. Across both test domains and two evaluation protocols, Mover360 outperforms strong baselines for perspective editing, insertion, and inpainting in reconstruction fidelity, semantic consistency, and distributional quality. Code and our benchmark dataset are available at https://zhonghaoyi.github.io/Mover360/.