1.cd vamr_submit
2.conda env create -f environment.yml && conda activate vamr
3.mkdir data and put dataset in this folder
4.cd code && python KLT_main.py and uncomment corresponding dataset.
5.For customized dataset,put the datahome folder into data
https://www.youtube.com/playlist?list=PLZHJcgEyXQg6rG42lscg3XZJcEEyUTxHO
We tested on a machine with CPU: 4*2.4 GHz, RAM: 16 GB, maximum number of used threads: 1.