Skip to content

Source paper

Geometry-Guided Reinforcement Learning for Multi-view Consistent 3D Scene Editing

Jiyuan Wang, Chunyu Lin, Lei Sun and 8 more

Jiyuan Wang, Chunyu Lin, Lei Sun, Zhi Cao, Yuyang Yin, Lang Nie, Zhenlong Yuan, Xiangxiang Chu, Yunchao Wei, Kang Liao, Guosheng Lin

arXiv 2603.03143 · 2026127 upvotes on Hugging Face

Research ideas from this paper (3)

Generated by AI. The paper’s authors did not write these.

AI summary

This paper introduces RL3DEdit, a novel reinforcement learning framework for single-pass 3D scene editing that achieves high-quality, multi-view consistent results without requiring explicit 3D-paired training data. It leverages a pre-trained 3D foundation model as a verifier to guide a multi-image 2D diffusion editor, effectively anchoring 2D editing priors onto a 3D-consistent manifold.

How it works (AI summary)

The framework utilizes the GRPO reinforcement learning algorithm to fine-tune a multi-image 2D diffusion model (FLUX-Kontext) by exploring diverse editing candidates and evaluating their multi-view consistency using geometry-aware rewards (depth confidence, point confidence, relative pose) extracted from a frozen 3D foundation model (VGGT).