Tensorflow implementation of paper Siamese Neural Networks for One-shot Image Recognition, Koch et al., 2015
To run, you'll first have to clone this repo and install the dependencies
git clone https://github.com/sonnguyentruong129/siamese-oneshot
cd siamese-oneshot
sudo pip install -r requirements.txt
Then you'll need to download the omniglot dataset and preprocess/pickle it with the load_data.py script.
git clone https://github.com/brendenlake/omniglot
python load_data.py --path <PATH TO THIS FOLDER>
Then you can run the jupyter notebook.
SiameseNet.ipynb
SiameseNet_VGG.ipynb
Note: SiameseNet implement the original architecture introduced in paper, SiameseNet_VGG fine tune pretrain VGG model
- Fine tuning pretrain with the original Siamese architecture
- Clean code, update Tensorflow 2.x