Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

JointEdit3D: Feed-Forward 3D Scene Editing in a Unified Latent Space

About

Existing 3D scene editing methods typically rely on per-scene optimization over explicit 3D representations or cascaded edit-and-reconstruct pipelines, resulting in high test-time cost, limited 3D awareness, and structural inconsistencies. To couple appearance synthesis and geometry prediction during editing, we build on a unified RGB-geometry reconstruction-generation latent space and adapt it to feed-forward 3D scene editing. The resulting framework, \textbf{JointEdit3D}, performs asymmetric latent inpainting by observing only a single edited RGB reference latent and generating the remaining RGB views and edited geometry latent under source-scene anchoring. JointEdit3D introduces a dedicated SceneAnchor Branch to inject source-scene structure without forcing direct copying, and adopts edit/background-aware losses to balance edited-region fidelity with unedited-content preservation. To address the lack of paired resources for standardized 3D scene editing evaluation, we introduce SceneEdit3D-15K, a dataset with 15K paired editing samples and renderer-provided 3D annotations, together with SceneEdit3D-Bench, a curated 100-sample benchmark. Experiments show that JointEdit3D improves edited-region quality and 3D structural completeness over prior baselines while maintaining competitive background preservation.

Xinnan Zhu, Ruijie Xu, Jiayu Ying, Daoguo Dong, Jiachen Xu, Yuan Xie, Xin Tan• 2026

Related benchmarks

TaskDatasetResultRank
DeletionSceneEdit3D-Bench
PSNR31.92
7
Deletion360-USID
PSNR18.57
7
3D Scene EditingSceneEdit3D-15K
Time (s)11.94
6
3D Scene EditingSceneEdit3D-Bench Edit (test)
PSNR (Delete)31.92
4
3D Scene EditingSceneEdit3D-Bench Background (test)
PSNR (Delete)32.62
4
Showing 5 of 5 rows

Other info

Follow for update