deep-hands

The aim of this project is to classify the user-shelf interaction. Three classes are considered:

Neutral	Positive	Negative

In order to generate *.npy files, you can run data.py script.
After this, you can use one of the following nets:

• CNN (train_cnn.py)
• CNN2 (train_cnn2.py)
• AlexNet (train_alexnet.py)
• CaffeNet (train_caffenet.py)

python data.py -images dataset/hands/ [-data data.csv]
python train_cnn.py
python train_cnn2.py
python train_alexnet.py
python train_caffenet.py

annotation-toolbox

This is a useful toolbox to annotate a set of images with different classes.

1. Positive (hand with product)
2. Negative (other)
3. Neutral (only hand)
4. Skip (bad image)

python annotation-toolbox.py -images dataset/hands/ [-data data.csv]

Python Environment Setup

sudo apt-get install python3-pip python3-dev python-virtualenv # for Python 3.n
virtualenv -p python3 venv
. venv/bin/activate

The preceding command should change your prompt to the following:

(venv)$

Install TensorFlow in the active virtualenv environment:

pip3 install --upgrade tensorflow-gpu # for Python 3.n and GPU

Install the others library:

pip3 install --upgrade keras scikit-learn scikit-image h5py

Author

• Daniele Liciotti | GitHub