SE360: Semantic Edit in 360$^\circ$ Panoramas via Hierarchical Data Construction
This work addresses a domain-specific problem for applications in virtual reality or panoramic imaging, representing an incremental advancement by adapting existing editing techniques to a challenging format.
The paper tackles the problem of instruction-based image editing for 360° panoramas, which often yields implausible results, by proposing SE360, a framework that uses a novel data generation pipeline and a Transformer-based diffusion model to achieve improved visual quality and semantic accuracy compared to existing methods.
While instruction-based image editing is emerging, extending it to 360$^\circ$ panoramas introduces additional challenges. Existing methods often produce implausible results in both equirectangular projections (ERP) and perspective views. To address these limitations, we propose SE360, a novel framework for multi-condition guided object editing in 360$^\circ$ panoramas. At its core is a novel coarse-to-fine autonomous data generation pipeline without manual intervention. This pipeline leverages a Vision-Language Model (VLM) and adaptive projection adjustment for hierarchical analysis, ensuring the holistic segmentation of objects and their physical context. The resulting data pairs are both semantically meaningful and geometrically consistent, even when sourced from unlabeled panoramas. Furthermore, we introduce a cost-effective, two-stage data refinement strategy to improve data realism and mitigate model overfitting to erase artifacts. Based on the constructed dataset, we train a Transformer-based diffusion model to allow flexible object editing guided by text, mask, or reference image in 360$^\circ$ panoramas. Our experiments demonstrate that our method outperforms existing methods in both visual quality and semantic accuracy.