Study of Eye Socket Recognition
Harisu Abdullahi SHEHU1*, Ibrahim Furkan INCE2, Faruk BULUT3
1 School of Engineering and Computer Science, Victoria University of Wellington, Wellington, NEW ZEALAND
harisushehu [at] ecs (dot) vuw (dot) ac (dot) nz, ORCID: 0000-0002-9689-3290
2 Department of Digital Game Design, Nisantasi University, Istanbul, TURKEY
ibrahim (dot) ince [at] nisantasi( dot) edu (dot) tr, ORCID: 0000-0003-1570-875X
3 Department of Computer Engineering, Istanbul Arel University, Istanbul, TURKEY
bulutfaruk[at] gmail (dot) com, ORCID: 0000-0003-2960-8725
* Corresponding author
Codes
For the eye socket recognition work, the link to the implemented Python codes with no data is as:
https://drive.google.com/file/d/1bHz3oaH6Btkn2SFNKVpB0dwAE4rcrUQh/view?usp=sharing
The file EyeSocketRecognition.pde needs to be run to generate the Gabor images and the results of the proposed method (without deep learning).
The Python code model.py will generate results obtained by the Proposed method + DL (deep learning).
The Python codes InceptionV3.py, VGG19.py, and ResNet50.py will generate results obtained by the state-of-the-art methods.
Datasets
The CK+ dataset can be downloaded from: http://www.jeffcohn.net/Resources/
You can also download the datasets used in our study:
[28] URL: https://104fashion.com/pictures-of-before-and-after-makeup/, received date: 01/05/2023
[29] URL: https://pictures.dxtr.com/The-Face-We-Make/n-D8Ps72/TFWM-Cropped/, received date: 01/05/2023
[30] https://www.flickr.com/photos/thefacewemake/albums
[31] BioID, https://www.bioid.com/About/BioID-Face-Database
Flickr Dataset
You can also the FLICKR dataset including 30 individuals from here.
Running Environment
The proposed method is implemented in Python and Java using the processing software. All java codes are implemented in the Processing 2.2.1 environment, therefore researchers should download the processing software to replicate our results.
Below is a link to download the Processing software for different operating systems:
Windows 32 bit: http://download.processing.org/processing-2.2.1-windows32.zip
Windows 64 bit: http://download.processing.org/processing-2.2.1-windows64.zip
Linux 32 bit: http://download.processing.org/processing-2.2.1-linux32.tgz
Linux 64 bit: http://download.processing.org/processing-2.2.1-linux64.tgz
Mac OS X: http://download.processing.org/processing-2.2.1-macosx.zip
For futher studies please contact with the authors. Their e-mail addresses are written above.
You can cite the reference in your studies:
Harisu Abdullahi SHEHU, Ibrahim Furkan INCE, Faruk BULUT, (2023), Enhancing Eye Socket Recognition Performance using Inverse Histogram Fusion Images in Gabor Transform, pp 1-15.