Benchmarking gradient-free optimizers for 3D performance capture in the nevergrad platform

Alexandros Doumanoglou, Nikolaos Zioulis, Vladimiros Sterzentsenko, Antonis Karakottas, Dimitrios Zarpalas, Petros Daras · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2021

In this paper we document our work that extends the nevergrad platform by adding a real-world benchmark at the intersection of computer vision and graphics: 3D Performance Capture. Detailed documentation can be found at vcl3d.github.io/nevergrad/

Read the paper · More papers on PaperTik