An improvement of the implementation from @gliese581gg, with added training, testing and video parsing. We also used the VOC tools to parse the VOC dataset from @mprat.

Paper: https://arxiv.org/abs/1506.02640

I know, I know: the code needs cleaning, bruh.

• Tensorflow
• OpenCV2

cd data_parsing

python data_parsing/setup.py install

Download YOLO weight file from: [https://drive.google.com/file/d/0B2JbaJSrWLpza08yS2FSUnV2dlE/view?usp=sharing].

Put the 'YOLO_small.ckpt' in the 'weight' folder of downloaded code.

cd data_parsing

python setup.py develop --uninstall

python network/YOLO_small_tf.py -fromfile "name of input file" -tofile_img "name of output file"

python network/YOLO_small_tf.py -video "name of input file" -tofile_vid "name of output file"

Refeer to the LICENSE files of both data_parsing and network.