3D Shape Editing
Overview
3D shape editing studies how to make targeted changes to a digital object while keeping its intended structure and other parts intact. Methods use mesh deformation, implicit or point-based representations, Gaussian splats, and generative models guided by text, images, sketches, masks, or semantic parts. Some rely on optimization or inversion for precise control, while feed-forward models aim to make edits faster and consistent across viewpoints. A central challenge is balancing strong, localized edits with preservation of unedited geometry, appearance, and identity; reliable evaluation of complex iterative edits remains difficult.
Latest papers
53 papers6 papers in the last four weeks, up 100% on the four weeks before. 0.0% of all new papers.