G3Flow: Generative 3D Semantic Flow for Pose-aware and Generalizable Object Manipulation

TL;DR

G3Flow introduces a real-time semantic flow framework that enhances 3D robotic manipulation by integrating foundation models, leading to improved success rates and generalization in complex tasks.

Contribution

The paper presents G3Flow, a novel framework combining 3D generative models, foundation models, and pose tracking to achieve semantic understanding without manual annotations.

Findings

Achieves up to 68.3% success in terminal-constrained tasks.

Attains 50.1% success in cross-object generalization.

Outperforms existing methods across five simulation tasks.

Abstract

Recent advances in imitation learning for 3D robotic manipulation have shown promising results with diffusion-based policies. However, achieving human-level dexterity requires seamless integration of geometric precision and semantic understanding. We present G3Flow, a novel framework that constructs real-time semantic flow, a dynamic, object-centric 3D semantic representation by leveraging foundation models. Our approach uniquely combines 3D generative models for digital twin creation, vision foundation models for semantic feature extraction, and robust pose tracking for continuous semantic flow updates. This integration enables complete semantic understanding even under occlusions while eliminating manual annotation requirements. By incorporating semantic flow into diffusion policies, we demonstrate significant improvements in both terminal-constrained manipulation and cross-object generalization. Extensive experiments across five simulation tasks show that G3Flow consistently outperforms existing approaches, achieving up to 68.3% and 50.1% average success rates on terminal-constrained manipulation and cross-object generalization tasks respectively. Our results demonstrate the effectiveness of G3Flow in enhancing real-time dynamic semantic feature understanding for robotic manipulation policies.