Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
normalization_layer
protos
scripts_prepare_training_data
.gitattributes
LICENSE
README.md
run_SINT_OTB.py

README.md

Siamese Instance Search for Tracking

Project page: https://staff.fnwi.uva.nl/r.tao/projects/SINT/SINT_proj.html

If you find our work useful, please consider citing:

@inproceedings{tao2016sint,
  title={Siamese Instance Search for Tracking},
  author={Tao, Ran and Gavves, Efstratios and Smeulders, Arnold W M},
  booktitle={CVPR},
  year={2016}
}


Prerequisites: Fast-RCNN https://github.com/rbgirshick/fast-rcnn (The region-of-interest pooling layer is used in our implementation) or Caffe (with the region-of-interest pooling layer added).


Trained caffe model: http://isis-data.science.uva.nl/rantao/SINT_similarity.caffemodel

[Run the tracker] After adding the normalization layer into your Fast-RCNN or Caffe, and modifying the path where necessary, you should be able to reproduce our experiments on OTB benchmark with the provided model by running 'run_SINT_OTB.py'.

[Training data] The scripts in 'scripts_prepare_training_data' shows how to process ALOV (http://alov300pp.joomlafree.it/dataset-resources.html) to prepare training data to train the Siamese network.


You can find more information, such as the result files on OTB, on the project page https://staff.fnwi.uva.nl/r.tao/projects/SINT/SINT_proj.html