RFTrans: Leveraging Refractive Flow of Transparent Objects for Surface Normal Estimation and Manipulation

TL;DR

RFTrans introduces a novel RGB-D-based approach leveraging refractive flow to improve surface normal estimation and manipulation of transparent objects, effectively bridging the sim-to-real gap and achieving high grasp success rates.

Contribution

The paper presents RFTrans, a new method that uses refractive flow as an intermediate representation for better geometry estimation of transparent objects, enabling effective robotic manipulation.

Findings

Outperforms baseline in synthetic and real-world tests

Achieves 83% success rate in real robot grasping

Uses synthetic data for training with successful sim-to-real transfer

Abstract

Transparent objects are widely used in our daily lives, making it important to teach robots to interact with them. However, it's not easy because the reflective and refractive effects can make depth cameras fail to give accurate geometry measurements. To solve this problem, this paper introduces RFTrans, an RGB-D-based method for surface normal estimation and manipulation of transparent objects. By leveraging refractive flow as an intermediate representation, the proposed method circumvents the drawbacks of directly predicting the geometry (e.g. surface normal) from images and helps bridge the sim-to-real gap. It integrates the RFNet, which predicts refractive flow, object mask, and boundaries, followed by the F2Net, which estimates surface normal from the refractive flow. To make manipulation possible, a global optimization module will take in the predictions, refine the raw depth, and construct the point cloud with normal. An off-the-shelf analytical grasp planning algorithm is followed to generate the grasp poses. We build a synthetic dataset with physically plausible ray-tracing rendering techniques to train the networks. Results show that the proposed method trained on the synthetic dataset can consistently outperform the baseline method in both synthetic and real-world benchmarks by a large margin. Finally, a real-world robot grasping task witnesses an 83% success rate, proving that refractive flow can help enable direct sim-to-real transfer. The code, data, and supplementary materials are available at https://rftrans.robotflow.ai.