Improving low-cost sensors to map sharp V-shaped surfaces more accurately
Reliability-Aware Reconstruction of V-Shaped Surfaces with a Low-Cost Multi-Zone SPAD Sensor
Computer Vision and Pattern Recognition
Summary
Measuring the shape of surfaces with sharp corners is hard for simple, cheap sensors because their readings can be sparse and unreliable. The authors developed a method to judge how trustworthy each measurement is by looking at how confident the sensor is and the quality of the detected signal. They then use this reliability information to improve the 3D shape reconstruction of surfaces with V-shaped edges, such as corners at different angles. Their approach reduces errors in angle and depth measurements compared to previous methods.
What this means in practice
- •For robotics engineers: Improve robot understanding of sharp corner shapes using low-cost depth sensors for better navigation and manipulation.
- •For industrial inspection teams: Use enhanced depth reconstruction to detect and measure V-shaped features and edges in manufactured parts more reliably with affordable sensors.
Authors
Yohanssen Pratama
Abstract
Low-cost multi-zone single-photon avalanche diode (SPAD) sensors provide compact time-of-flight measurements but are limited by sparse sampling and unreliable depth estimates near geometric discontinuities. This letter presents a reliability-aware reconstruction framework for V-shaped surfaces using two complementary TMF8821 acquisition configurations. Point-wise reliability is learned from sensor confidence, transient peak amplitude, and peak sharpness, and is incorporated into a graph-based depth optimization. The reconstructed geometry is evaluated by robust two-plane fitting and a perpendicular cross-section. Experiments on 60-degree, 90-degree, and 120-degree targets reduce the mean V-angle error from 11.21 degrees to 9.66 degrees (13.8%). For the representative 90-degree target, the combined point-to-plane RMSE decreases from 14.56 mm to 13.65 mm.