Vehicle Detection in Aerial Imagery (VEDAI) : a benchmark

Sebastien Razakarivony and Frederic Jurie, 2014

Introduction

VEDAI is a dataset for Vehicle Detection in Aerial Imagery, provided as a tool to benchmark automatic target recognition algorithms in unconstrained environments. The vehicles contained in the database, in addition of being small, exhibit different variabilities such as multiple orientations, lighting/shadowing changes, specularities or occlusions. Furthermore, each image is available in several spectral bands and resolutions. A precise experimental protocol is also given, ensuring that the experimental results obtained by different people can be properly reproduced and compared. We also give the performance of some baseline algorithms on this dataset, for different settings of these algorithms, to illustrate the difficulties of the task and provide baseline comparisons.
image

Database citation

Please, cite Razakarivony and Jurie [1] when using the database.

Download and Copyrights

512x512 images: [annotations] , [part1], [part2]
1024x1024 images: [annotations] , [part1], [part2], [part3], [part4], [part5]
development kit: [devkit]
Terms of Use: [terms]

Contact

Kindly contact us (frederic.jurie@unicaen.fr) for questions, comments etc. about the database.

References

[1] Vehicle Detection in Aerial Imagery: A small target detection benchmark., Sébastien Razakarivony and Frédéric Jurie, Journal of Visual Communication and Image Representation, 2015 [Paper]