----------------------------------
Konstantinos Papoutsakis
----------------------------------
Computer Vision & Robotics Lab
Institute of Computer Science (ICS)
Foundation of Research & Technology (FORTH)
Crete, Greece

www.ics.forth.gr/~papoutsa
www.ics.forth.gr/cvrl/evaco
----------------------------------
Last edited: 10 March 2017
----------------------------------

80-pair Dataset 
The dataset was generated by Zhuwen LI and colleagues, was introduced in [1]:
[1] Guo, Jiaming, et al. "Video co-segmentation for meaningful action extraction." Proceedings of the IEEE international conference on computer vision. 2013.

The videos, ground truth files of the dataset and the source code for their method can be downloaded from http://www.lizhuwen.com/ 
or directly from http://online.ece.nus.edu.sg/VideoDataset/iccv13_coseg_dataset.zip

We provide the commputed data BoW-based features per sequence of the 80-pair dataset, used in our work. 
Read the following description and the paper for more details on the provided data.
--------------------------------------------------------------------------------------------------------------------------------------------------------
--> data folder
The video-based features per pair of sequences of the 80-Pair dataset. We have processed the original videos provided in the dataset using the software 
for extracting Dense Point Tracking provided by Thomas Brox in http://lmb.informatik.uni-freiburg.de/resources/software.php and the used 
the code provided by the authors to extract MBH-based descriptors (one feature vector per frame), in order to acquire the same type of data
that is also used in their methodology [1].
--------------------------------------------------------------------------------------------------------------------------------------------------------

--------------------------------------------------------------------------------------------------------------------------------------------------------
--> gt_txt folder
The ground truth data for each pair of sequences, as provided in the original dataset http://online.ece.nus.edu.sg/VideoDataset/iccv13_coseg_dataset.zip
--------------------------------------------------------------------------------------------------------------------------------------------------------