NLiPsCalib: An Efficient Calibration Framework for High-Fidelity 3D Reconstruction of Curved Visuotactile Sensors
Xuhao Qin, Feiyu Zhao, Yatao Leng, Runze Hu, Chenxi Xiao
AI summary
Problem
Calibrating curved visuotactile sensors traditionally requires expensive, labor-intensive specialized hardware or complex simulations to map non-uniform internal illumination to surface normals, hindering rapid sensor customization.
Approach
The framework adapts Near-Light Photometric Stereo to model the sensor's internal lighting, enabling direct geometry estimation from casual presses on everyday objects to train a real-time neural network.
Key results
- Hardware-free calibration using Near-Light Photometric Stereo
- NLiPsTac modular sensor platform with controllable LEDs
- NLiPsNet lightweight network for real-time normal inference
- Reconstruction accuracy matching state-of-the-art device-based methods
Why it matters
Drastically reduces the cost and expertise required for tactile sensor development, making high-fidelity 3D reconstruction accessible to a broader robotics community.
Abstract
Recent advances in visuotactile sensors increas- ingly employ biomimetic curved surfaces to enhance senso- rimotor capabilities. Although such curved visuotactile sen- sors enable more conformal object contact, their perceptual quality is often degraded by non-uniform illumination, which reduces reconstruction accuracy and typically necessitates calibration. Existing calibration methods commonly rely on customized indenters and specialized devices to collect large- scale photometric data, but these processes are expensive and labor-intensive. To overcome these calibration challenges, we present NLiPsCalib, a physics-consistent and efficient calibra- tion framework for curved visuotactile sensors. NLiPsCalib integrates controllable near-field light sources and leverages Near-Light Photometric Stereo (NLiPs) to estimate contact geometry, simplifying calibration to just a few simple contacts with everyday objects. We further introduce NLiPsTac, a controllable-light-source tactile sensor developed to validate our framework. Experimental results demonstrate that our approach enables high-fidelity 3D reconstruction across diverse curved form factors with a simple calibration procedure. We emphasize that our approach lowers the barrier to developing customized visuotactile sensors of diverse geometries, thereby making visuotactile sensing more accessible to the broader community.