Perceiving physical contact is fundamental to dexterous manipulation. While robots often rely on dedicated hardware tactile sensors, humans exhibit a remarkable ability to infer contact by integrating visual information with an innate sense of their body's pose and movement. Inspired by this embodied perceptual skill, we investigate whether a robot can learn to infer contact from vision, an approach that also offers a scalable alternative to tactile hardware specifically for binary contact estimation, which faces practical challenges in cost, fragility, and integration
We present NoContactNoWorries, a transformer-based multimodal framework that fuses RGB-D vision with the robot's proprioception to infer binary contact states as a pseudo-tactile signal for hand-object interactions. We validate by training a single contact prediction model on multiple objects and show that the inferred contact signal supports downstream reinforcement learning agents for in-hand object reorientation, generalizing to novel objects. Experiments in both simulation and on a real-world robot validate our approach, highlighting the feasibility of inferring contact from vision and proprioception.
Our framework combines RGB-D observations with robot proprioception to predict binary contact states between the robotic hand and the manipulated object. The predicted contact signal serves as a pseudo-tactile representation and is provided to a downstream reinforcement learning policy for dexterous in-hand manipulation.
Training object used for policy learning and base evaluation.
Real-world manipulation of an unseen Hexagonal Prism object.
Quantitative comparison of NoContactNoWorries on simulation and real-world dexterous objects.
If you find this work useful in your research, please consider citing:
@misc{patil2026nocontactnoworriesestimatingcontactvision,
title={NoContactNoWorries: Estimating Contact through Vision and Proprioception for In-Hand Dexterous Manipulation},
author={Soham Patil and Avirup Das and Sourabh Bhosale and Spandan Roy},
year={2026},
eprint={2606.24450},
archivePrefix={arXiv},
primaryClass={cs.RO},
url={https://arxiv.org/abs/2606.24450}
}