Local Changes Invariant Image Matching Database

Goals
The goal of this database is to provide researchers a collection of retouched images for one-to-one image matching task. With the development of image editing softwares, content based image matching is facing the issue of local changes brought by human, because the local changes (noise) can dramatically affect the matching performance of keypoint based features. We address this problem in our research and wish better results can be achieved. The query set is made manually with the help from laboratory members by using popular image editing softwares. Top 5,000 images in the candidate set are the ground truth of the images in the query set respectively. Most of the candidate images are from SUN databse.

Citation
If you find this dataset useful, please cite one of the following papers:
[1] @conference{zhang2015compressive,
title={Compressive Image Retrieval with Modified Images},
author={Zhang, Chao and Akashi, Takuya},
booktitle={10th Asian Control Conference (ASCC)},
year={2015}
}
[2] @article{zhang2015matching,
title={High-speed and Local-changes Invariant Image Matching},
author={Zhang, Chao and Akashi, Takuya},
journal={IEICE Transactions on Information and Systems},
volume={98},
number={11},
year={2015},
publisher={IEICE}
}

Copyright

THIS DATA SET IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.

The images provided above may have certain copyright issues. We take no guarantees or responsibilities, whatsoever, arising out of any copyright issue. Please use it at your own risk.