Oxford Buildings Benchmark

This is a repository that contains a dump of the Oxford VGG building dataset benchmark data, with the evaluation script.

This is more or less a vis-a-vis clone of the ground truth data, only published to be open about any potential disputes as the benchmark script has changed.

Computing the average precision with Python

• python compute_ap.py christ_church_1 rank_list.txt for the first Christ Church query.

Computing the average precision with original C++ code

This is only provided as a baseline for comparison - we use the Python evaluation script for all of the benchmark results we publish.

• Compile the compute_ap.cpp file, using (on linux) g++ -O compute_ap.cpp -o compute_ap.
• To compute the average precision for a ranked list, rank_list.txt, one runs ./compute_ap christ_church_1 rank_list.txt, for the first Christ Church query, etc.

Acknowledgements

This data has been provided by the Oxford Visual Geometry Group.

Philbin, J. , Chum, O. , Isard, M. , Sivic, J. and Zisserman, A. Object retrieval with large vocabularies and fast spatial matching - Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2007)