Adrian-Stefan Ungureanu's Projects
Demonstration of FusionNet for touchless palmprint and finger texture recognition. Based on the source code for the 2019 IEEE CIVEMSA paper "Touchless palmprint and finger texture recognition: A Deep Learning fusion approach"
Face recognition using Tensorflow
Facenet implementation by Keras2
Keras model trained using semi-hard triplet Loss (tensorflow function) on MNIST
Simple script to label an image with a finite number of landmarks. The output is an XML script containing the (x,y) coordinates of the points.
Using 3D coordinates relative to Leap Motion, project the points onto a 2D image plane
Learn OpenCV : C++ and Python Examples
Machine Learning Experiments and Work
Large palmprint database acquired in unconstrained conditions, allowing the training of data-intensive models for palmprint ROI extraction/detection
Various palmprint feature extraction techniques: CompCode, Local Tetra Patterns, RLOC
Code associated with ICCE 2018 paper "Deep learning for hand segmentation in complex backgrounds"
A Python implementation of PCANet https://arxiv.org/abs/1404.3606
Example of using RAPT to package Pygame(_sdl2) games.
SegNet including indices pooling for Semantic Segmentation with tensorflow and keras
Few Shot Learning by Siamese Networks, using Keras.
Keras Implementation for Stacked Hourglass Network
A crash course in six episodes for software developers who want to become machine learning practitioners.
Scripts used for the experiments carried out in the paper "Unconstrained palmprint as a smartphone biometric", published by IEEE Transactions on Consumer Electronics (2017)