Geometric Image Editing via Effects-Sensitive In-Context Inpainting with Diffusion Transformers
Preprint
Abstract
Recent advances in diffusion models have significantly improved image editing. However, challenges persist in handling geometric transformations, such as translation, rotation, and scaling, particularly in complex scenes. Existing approaches suffer from two main limitations: (1) difficulty in achieving accurate geometric editing of object translation, rotation, and scaling; (2) inadequate modeling of intricate lighting and shadow effects, leading to unrealistic results. To address these issues, we propose GeoEdit, a framework that leverages in-context generation through a diffusion transformer module, which integrates geometric transformations for precise object edits. Moreover, we introduce Effects-Sensitive Attention, which enhances the modeling of intricate lighting and shadow effects for improved realism. To further support training, we construct RS-Objects, a large-scale geometric editing dataset containing over 120,000 high-quality image pairs, enabling the model to learn precise geometric editing while generating realistic lighting and shadows. Extensive experiments on public benchmarks demonstrate that GeoEdit consistently outperforms state-of-the-art methods in terms of visual quality, geometric accuracy, and realism.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- ORIDa: Object-centric Real-world Image Composition Dataset2025
- ObjectMover: Generative Object Movement with Video Prior2025
- Diffusion Handles Enabling 3D Edits for Diffusion Models by Lifting Activations to 3D2024
- Image Sculpting: Precise Object Editing with 3D Geometry Control2024
- DragDiffusion: Harnessing Diffusion Models for Interactive Point-Based Image Editing2024
- ObjectStitch: Object Compositing with Diffusion Model2023
- Paint by Example: Exemplar-based Image Editing with Diffusion Models2023
- Images Speak in Images: A Generalist Painter for In-Context Visual Learning2023
- DoveNet: Deep Image Harmonization via Domain Verification2020
- Deep Image Harmonization2017