Skip to content

zeynepnurozturk/CS484-Image-Analysis

Repository files navigation

Ahmet Furkan Biyik - 21501084
Zeynep Nur Ozturk - 21501472
Ebru Kerem - 21501859
09.01.2020

==================================================================
To run the program the following must be installed: python, pip
The following command installs the required libraries: pip install -r requirements.txt
==================================================================
Preprocessed images were saved to "train.mat" and "test.mat" files.
You can remove this files and preprocess data provided in course page:
"http://www.cs.bilkent.edu.tr/~saksoy/courses/cs484-Fall2019/src/cs484_project_data.tar.gz"
You can change data paths in "main.py".
Prerocess operation creates exactly same "train.mat" and "test.mat" files.
Prerocess operation may take more than 12 hours.
==================================================================
Trained network will be saved as "network.pth" file.
You can change network's parametes in "main.py" file.
If you want to change parameters, you need to delete "network.pth" file.
==================================================================
You can run the program using "python main.py"
==================================================================
Results of the program will be saved into results folder.
"boxes_*.png" files are candidate window examples.
"local_*.png" files are localization result examples.
In localization result examples, red is given window, blue is predicted window.
"result.txt" file contains statistics about network test.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages