IEEE Transactions on Pattern Analysis and Machine Intelligence

High-speed 3D shape measurement with structured light methods: A review

Monocular depth prediction plays a crucial role in understanding 3D scene geometry. Although recent methods have achieved impressive progress in evaluation metrics such as the pixel-wise relative error, most methods neglect the geometric constraints in the 3D space. In this work, we show the importance of the high-order 3D geometric constraints for depth prediction. By designing a loss term that enforces one simple type of geometric constraints, namely, virtual normal directions determined by randomly sampled three points in the reconstructed 3D space, we can considerably improve the depth prediction accuracy. Furthermore, we can not only predict accurate depth but also achieve high-quality other 3D information from the depth without retraining new parameters, Significantly, the byproduct of this predicted depth being sufficiently accurate is that we are now able to recover good 3D structures of the scene such as the point cloud and surface normal directly from the depth, eliminating the necessity of training new sub-models as was previously done. Experiments on two challenging benchmarks: NYU Depth-V2 and KITTI demonstrate the effectiveness of our method and state-of-the-art performance.

/pdf/enforcing-geometric-constraints-of-virtual-normal-for-depth-dskrdmury1.pdf

Enforcing Geometric Constraints of Virtual Normal for Depth Prediction

Absolute phase retrieval methods for digital fringe projection profilometry: A review

Status, challenges, and future perspectives of fringe projection profilometry

A flexible phase-shifting method with absolute phase marker retrieval

Accurate and robust phase unwrapping is an important procedure for three dimensional (3D) profilometry measurement. The mask cut phase-unwrapping algorithm merges the characteristics of Goldstein's branch cut and reliability-guided path-following algorithms, so that it has higher accuracy and stability. However, it cannot handle discontinuous phase regions because it completely relies on the phase quality map to guide the placement of mask cuts, thereby easily causing phase error propagation. To overcome these drawbacks, we propose a new phase-unwrapping method that merges the residue check principle and a Laplace phase derivative variance quality map that we developed. First, the entire phase was divided into several isolated regions in accordance with the quality map. Then, each pixel of the absolute phase marker line was taken as the starting point to unwrap the discontinuous phase regions based on reliability guidance. In addition, a new list-trimming algorithm was employed to guarantee a speedy phase-unwrapping procedure. The entire phase unwrapping and accurate 3D measurement of discontinuous objects was successfully completed. The simulated and experimental data both demonstrate the validity of the proposed phase-unwrapping algorithm.

Reliability-guided phase-unwrapping algorithm for the measurement of discontinuous three-dimensional objects

A calibration strategy for vision-guided robot assembly system of large cabin

Vision measurement plays a significant role in robot automatic assembly. In this study, a new measurement method for robot peg-in-hole prealignment via a combined two-level vision system is proposed. The assembly system and a global coordinate system calibration procedure based on a dynamic coordinate system are developed to construct an accurate transformation between the coordinate systems. The subsequent hole pose is obtained with the proposed image processing method and the hole edge matching method, followed by the 3-D hole reconstruction and spatial circle fitting algorithm. A low-cost pose determination is proposed based on hole edge point selection. The experiment and analysis indicate that the hybrid measurement system has high precision in the local hole pose and global robot pose measurement accuracy. The peg-in-hole assembly results verify the measurement and calibration accuracy sideways. Our hybrid measurement system enhances the vision measurement range, local accuracy, and robot automation.

Haihua Cui

Papers

A flexible phase-shifting method with absolute phase marker retrieval

Reliability-guided phase-unwrapping algorithm for the measurement of discontinuous three-dimensional objects

A calibration strategy for vision-guided robot assembly system of large cabin

A Measurement Method for Robot Peg-in-Hole Prealignment Based on Combined Two-Level Visual Sensors

Dual-camera-based method for identification and location of scattered self-plugging rivets for robot grasping