Scaling Versatile 3D Assets Editing with a Million-Scale Dataset

2026-09-28

AI Summary

This study addresses the challenges of training data scarcity, inadequate source-aware modeling, and the absence of standardized evaluation protocols in 3D asset editing by proposing Alchemy3D, a unified framework. Methodologically, we construct the first million-scale (1.25M) 3D editing dataset alongside an open-world benchmark, GEdit3D-Bench. The proposed architecture is built upon generative flow models, supporting multimodal text-image conditioning, few-step inference, and multi-view part segmentation transfer. Experimental results demonstrate that Alchemy3D significantly outperforms existing methods across core metrics, including editing fidelity, source consistency, and visual quality.

Abstract

Although recent 3D generative models produce increasingly realistic assets, controllable 3D asset editing remains challenging. Existing methods are limited by scarce training data, insufficient source-aware modeling, and a lack of practical evaluation protocols. To address these limitations, we present Alchemy3D, a unified framework for training and evaluating versatile 3D asset editors that covers data construction, model architecture, and benchmark evaluation. Specifically, we curate Alchemy3D-1M, a large-scale 3D editing dataset containing 1.25M assets and 1.38M editing pairs across seven editing types. On this data, we train a family of generative flow models for general-purpose 3D asset editing. The model family supports image- and text-conditioned editing, few-step inference, and transfer to multi-view 3D part segmentation. We further introduce GEdit3D-Bench, a large-scale, open-world benchmark with a multi-dimensional evaluation protocol. Across existing and newly introduced benchmarks, our method outperforms prior methods on most metrics of editing fidelity, source preservation, and visual quality.

Problem

• 3D asset editing
• training data scarcity
• source-aware modeling
• evaluation benchmark

Innovation

• 3D asset editing
• generative flow models
• large-scale dataset
• benchmark evaluation
• controllable generation