VM-NET in Tensorflow

Tensorflow implementation of SSPP-DAN: Deep Domain Adaptation Network for Face Recognition with Single Sample Per Person

Demo

Prerequisites

• Python 2.7
• Tensorflow 0.12
• OpenCV 2.4.9
• NumPy
• SciPy

Usage

First, download video dataset from "data/YoutubeID.txt." Then, extract audio features and video features from the videos.

Network structure

To train a model with downloaded dataset:

$ python train.py

To test with an existing model:

$ python test.py

Results