3DMorph: Single-Image-Guided Local 3D Shape Editing and Morphing

2026-06-05

AI Summary

Existing 3D shape editing methods often suffer from global modifications, strong domain dependence, complex user interaction, or mere appearance changes, lacking intuitive and efficient capabilities for local geometric editing. This work proposes the first training-free framework for single-image-guided local 3D editing, which estimates 2D–3D correspondences to automatically localize the editing region, transfer geometric deformations from the input image, preserve unedited regions, and support intermediate shape interpolation. To evaluate image-guided local 3D editing quality, we introduce the Delta3D benchmark. Experiments demonstrate that our method significantly outperforms existing generative and editing approaches on Delta3D, achieving—for the first time—accurate, high-quality local 3D geometric modifications directly from a single edited image.

Abstract

Despite recent progress in 3D generation, intuitive editing of existing shapes remains limited. Unlike images, which benefit from well-established inpainting tools, general 3D objects such as meshes still lack simple and effective methods for local shape editing. Existing approaches are often global, domain-specific, require complex user interaction, or focus on appearance (color and texture) rather than geometry. We introduce 3DMorph, a training-free framework for single-image-guided local 3D shape editing and morphing. Given an edited image showing a desired shape modification, our method automatically localizes the relevant 3D region and transfers 2D modifications to 3D while preserving unmodified areas. 3DMorph also enables intermediate shape generation between the original and edited objects, facilitating design exploration. To benchmark editing quality, we introduce Delta3D, an image-guided local 3D editing benchmark with paired ground-truth edits. Experimental results show that 3DMorph translates intuitive 2D edits into 3D, outperforming state-of-the-art generative and editing methods.

Problem

• 3D shape editing
• local editing
• geometry manipulation
• single-image guidance
• 3D morphing

Innovation

• 3D shape editing
• single-image guidance
• training-free
• local morphing
• geometry transfer