Efficient 3D triangulation in hardware for dense structure-from-motion in low-speed automotive scenarios

A Comprehensive Survey on Delaunay Triangulation: Applications, Algorithms, and Implementations Over CPUs, GPUs, and FPGAs

“…16b shows the corresponding speedup. Finally, although not based on Delaunay, it is worth mentioning the FPGA implementation of a triangulation function in [26], where the authors propose an architecture that employs linear triangulation with an inhomogeneous solution to the equation system. Specifically, [26] involved a comparison between an FPGA implementation and a software counterpart utilizing MATLAB, C++, and OpenCV.…”

“…Finally, although not based on Delaunay, it is worth mentioning the FPGA implementation of a triangulation function in [26], where the authors propose an architecture that employs linear triangulation with an inhomogeneous solution to the equation system. Specifically, [26] involved a comparison between an FPGA implementation and a software counterpart utilizing MATLAB, C++, and OpenCV. Notably, the FPGA implementation was able to process about 572 times more points per second with respect to the C++ software implementation and about 2122 times more than the MATLAB one.…”

“…In the literature, several algorithms have been developed in order to compute Delaunay triangulations, e.g., [14]- [16]. Moreover, in the last few years, the increasing demand for high-performance computing has led to the implementation of Delaunay triangulation algorithms on different hardware platforms, including CPUs [17]- [21], GPUs [22]- [25], and FPGAs [26]- [31]. These implementations aim to improve the computational efficiency and reduce the running time of triangulation algorithms, especially for large data sets.…”