Trans2Occ: Voxel Occupancy Estimation and Grasp for Transparent Objects from Simulation to Reality

2026-06-01

AI Summary

Transparent objects pose significant challenges for depth perception due to complex refraction and reflection effects, hindering accurate robotic grasping. This work proposes the first framework that directly predicts voxel-wise occupancy from a single-view RGB image, trained on large-scale synthetic data to learn domain-invariant geometric representations. The model enables effective sim-to-real transfer without fine-tuning, supporting occupancy-based rule-driven grasping strategies. Extensive experiments demonstrate high-fidelity 3D reconstruction and stable grasping performance in both simulated and real-world environments, validating the method’s robustness and practical applicability.

Abstract

Transparent objects remain challenging for robotic perception due to unreliable depth sensing caused by refraction and reflection. While prior approaches rely on multi-view reconstruction or depth completion, they are often difficult to scale or deploy in real-world robotic systems. In this paper, we present a practical framework for transparent object perception and manipulation based on single-view RGB input. Our approach predicts voxel-space occupancy directly from a single image, providing a geometry-aware representation that supports downstream robotic grasping. To enable large-scale training, we construct a simulation pipeline that generates paired RGB images and voxel occupancy annotations under diverse materials and lighting conditions. We demonstrate that the predicted occupancy representation is robust to domain shifts and transfers effectively from simulation to real-world robotic setups without fine-tuning. A simple rule-based grasping strategy built on top of the occupancy further achieves reliable grasp performance on transparent objects. Extensive experiments in both simulation and real-world environments show that our framework provides accurate 3D understanding and enables practical manipulation of transparent objects. These results suggest that single-view occupancy prediction offers a scalable and effective solution for transparent object perception in robotics.

Problem

• transparent objects
• robotic perception
• depth sensing
• voxel occupancy
• grasping

Innovation

• voxel occupancy estimation
• transparent object perception
• simulation-to-reality transfer
• single-view 3D reconstruction
• robotic grasping