Replication and Extension of the Research Paper: 'Dobs, Katharina & Yuan, Joanne & Martinez, Julio & Kanwisher, Nancy. (2022). Using deep convolutional neural networks to test why human face recognition works the way it does. 10.1101/2022.11.23.517478. '
Link to our GitHub Repository: https://github.com/hashtaghari/Understanding-Face-Recognition-Using-Deep-Convolutional-Neural-Networks.git
HOW TO RUN:
The folder 'Code and Datasets for Experiments' contains the code and the stimuli used to replicate the results of the paper, we tried to replicate the inferences in case the exact stimuli weren't availiable.
In order to run the experiments, you can download all the stimuli data in the link provided in the README.md file inside the folder 'Code and Datasets for Experiments'. After that, running the python file 'Analysis.ipynb' would show the results generated by our CNNs. You may need to download all the CNN models we trained. That would be availiable in the README.md file inside the folder 'models'.
Training CNNs:
The information and links to the datasets we used to train the CNNs is avaliable on the README.md file inside the folder 'Training CNNs'.
Note: All the datasets used to train the CNNs and the stimuli images used to replicate the experiments is linked in the repo. Please refer to our report on the results of our study and summary of the original work.