Bi-Adapt: Few-shot Bimanual Adaptation for Novel Categories of 3D Objects via Semantic Correspondence

Bi-Adapt: Few-shot Bimanual Adaptation for Novel Categories of 3D Objects via Semantic Correspondence

¹National University of Singapore ²Shanghai Qi Zhi Institute

³Peking University ⁴The University of Hong Kong
^*Equal Contribution ^†Corresponding Author

Abstract

Bimanual manipulation is imperative yet challenging for robots to execute complex tasks, requiring coordinated collaboration between two arms. However, existing methods for bimanual manipulation often rely on costly data collection and training, struggling to generalize to unseen objects in novel categories efficiently. In this paper, we present Bi-Adapt, a novel framework designed for efficient generalization for bimanual manipulation via semantic correspondence. Bi-Adapt achieves cross-category affordance mapping by leveraging the strong capability of vision foundation models. Fine-tuning with restricted data on novel categories, Bi-Adapt exhibits notable generalization to out-of-category objects in a zero-shot manner. Extensive experiments conducted in both simulation and real-world environments validate the effectiveness of our approach and demonstrate its high efficiency, achieving a high success rate on different benchmark tasks across novel categories with limited data.

Pipeline

(Left) We first train the Action Learning Network on the supporting set. The corresponding affordance and action distribution serve as our prior knowledge. (Middle) Then, we make contact points mapping from training categories to novel categories, leveraging the foundation model. (Lower Right) After that, the pre-trained network proposes actions based on mapped contact-point pairs on novel categories and fine-tuned with the interaction results. (Upper Right) Finally, the fine-tuned networks can facilitate manipulating unseen instances from novel categories with better performance.

More Visuals

Cross-category affordance generalization results

Each group shows a source image with contact points (★: first gripper, ★: second gripper) and a target image with inferred points. Highlight intensity indicates correspondence.

Adaptation Results

Visualization of manipulation results across tasks

Visualization of affordance comparison

Real Experiment Setup

(a) Robot Setup

(b) Objects

(a) We use two UFactory xArm6 for manipulation and two RealSense D435 cameras to cap- ture RGBD observations.
(b) We test on a diverse set of objects.

Real-world Demo