Paper 2510.15019
NANO3D: A Training-Free Approach for Efficient 3D Editing Without Masks
- Published
- Oct 2025
- Research lab
- Independent
- Citations
- 27
- GitHub
- 179 stars
01 In brief
Summary
Nano3D is a training-free framework for 3D object editing that supports removal, addition, and replacement tasks without requiring masks.
It integrates FlowEdit into the TRELLIS pipeline to perform localized edits guided by front-view renderings, and introduces region-aware merging strategies (Voxel/Slat-Merge) to preserve structural fidelity in unedited regions.
The framework uses a voxel-wise XOR operation to detect edited regions and merges them back into the source object.
Experiments show Nano3D outperforms baselines like Tailor3D, Vox-E, and TRELLIS in structure consistency, semantic alignment, and generation quality.
The authors also construct Nano3D-Edit-100k, a large-scale dataset with over 100,000 editing pairs, using a pipeline that samples images, generates instructions via Qwen2.5-VL, and edits with Nano Banana or Flux-Kontext.
The dataset shows better semantic alignment than 3D-Alpaca.
Limitations include support only for localized edits and constraints from TRELLIS's VAE reconstruction loss.
02 From the paper
Abstract
3D object editing is essential for interactive content creation in gaming, animation, and robotics, yet current approaches remain inefficient, inconsistent, and often fail to preserve unedited regions. Most methods rely on editing multi-view renderings followed by reconstruction, which introduces artifacts and limits practicality. To address these challenges, we propose Nano3D, a training-free framework for precise and coherent 3D object editing without masks. Nano3D integrates FlowEdit into TRELLIS to perform localized edits guided by front-view renderings, and further introduces region-aware merging strategies, Voxel/Slat-Merge, which adaptively preserve structural fidelity by ensuring consistency between edited and unedited areas. Experiments demonstrate that Nano3D achieves superior 3D consistency and visual quality compared with existing methods. Based on this framework, we construct the first large-scale 3D editing datasets Nano3D-Edit-100k, which contains over 100,000 high-quality 3D editing pairs. This work addresses long-standing challenges in both algorithm design and data availability, significantly improving the generality and reliability of 3D editing, and laying the groundwork for the development of feed-forward 3D editing models. Project Page:https://jamesyjl.github.io/Nano3D