Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation

TL;DR

Lift3D introduces a novel framework that enhances 2D large-scale pretrained models with 3D representations, enabling robust robotic manipulation by combining implicit and explicit 3D features.

Contribution

The paper presents a new method to adapt 2D foundation models for 3D robotic tasks through a task-aware autoencoder and a model-lifting strategy, addressing data scarcity and spatial information loss.

Findings

Lift3D outperforms previous methods in simulation benchmarks.

It demonstrates strong real-world manipulation capabilities.

The approach effectively integrates 3D information into 2D models.

Abstract

3D geometric information is essential for manipulation tasks, as robots need to perceive the 3D environment, reason about spatial relationships, and interact with intricate spatial configurations. Recent research has increasingly focused on the explicit extraction of 3D features, while still facing challenges such as the lack of large-scale robotic 3D data and the potential loss of spatial geometry. To address these limitations, we propose the Lift3D framework, which progressively enhances 2D foundation models with implicit and explicit 3D robotic representations to construct a robust 3D manipulation policy. Specifically, we first design a task-aware masked autoencoder that masks task-relevant affordance patches and reconstructs depth information, enhancing the 2D foundation model's implicit 3D robotic representation. After self-supervised fine-tuning, we introduce a 2D model-lifting strategy that establishes a positional mapping between the input 3D points and the positional embeddings of the 2D model. Based on the mapping, Lift3D utilizes the 2D foundation model to directly encode point cloud data, leveraging large-scale pretrained knowledge to construct explicit 3D robotic representations while minimizing spatial information loss. In experiments, Lift3D consistently outperforms previous state-of-the-art methods across several simulation benchmarks and real-world scenarios.