Benchmarking Pose Estimation Algorithms

Download all files and run “Batch_Benchmark.m” in MATLAB to benchmark camera pose estimation algorithms

• 8-point algorithm
• Kneip (5-point)
• Kukelova (5-point)
• Nister (5-point)
• Li & Hartley (5-point)
• Stewenius (5-point)
• QuEst (5-point)

on real world image datasets

• KITTI
• TUM
• ICL
• NAIST.

You need a 64-bit Windows machine to be able to benchmark all algorithms. Also, MATLAB’s Computer Vision System Toolbox and Statistics and Machine Learning Toolbox are required. See the Readme file for more details.

You can download the files individually from the Google Drive folder (right window), or you can download all files in the .zip format directly from here.

(A C++ implementation of the algorithm will be made available soon.)