This is a translation of ryanjay0/miles-deep in tf.keras.
This is just a model translation without any feature. The model is translated using microsoft/MMdnn.
The accuracy has not been verified since the data used to train the model is not available.
The description below is slightly modified from the original repository.
Using a deep convolutional neural network with residual connections, Miles Deep quickly classifies each second of a pornographic video into 6 categories based on sexual act with 95% accuracy.
- blowjob_handjob
- cunnilingus
- other
- sex_back
- sex_front
- titfuck
- apt install ffmpeg
- pip install tensorflow
- pip install opencv-python
- Put a short video with file name vid.mp4
- Run python test.py
It will convert the video into 1 image/second and print out only the labels (>0.7) ignoring "other" label.