Main Focus:
- Predict testing data genre based on the training data.
More detailed information can be found on the program source code or by clicking here.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
-
Navigate to the directory where is the root of this project folder.
-
Still in the directory where is the root of this project, install all its dependencies.
pip3 install -r requirements.txt
Dependencies are all listed in
requirements.txt
. To re-generate this file (after you've installed new packages), simply you can it directly on the file. If you have a problem installing the dependencies, please install python system package first. -
All the trailer videos is in MovieScope_DataSet directory.
-
Unspecific videos genre can be cleaned with clean.py
-
Limit to 1 film per trailer can be cleaned using extra_clean.py
The application is console-based which can be executed from the terminal. The general steps are as follows:
-
Execute create_data.py using genre e.g create_data.py [extractor]
**vgg16 or resnet
-
create_model.py [model_name_train]
**mlp, spatial, or lstm
-
test_model.py [video/test] [....] [....]
**If video then [video_Path] [model_filename]
**If test then [model_name] [extractor_name]
-
The extracted feature & model will be saved on data directory
- OpenCV - 3.4.0.12
- Keras - 2.1.6
- Tensorflow - 1.8.0
- Adityo Anggraito - Computer Science Student University of Indonesia - GitHub
- Bryanza Novirahman - Computer Science Student University of Indonesia - GitHub
- Muhammad Faiz - Computer Science Student University of Indonesia - GitHub
- Muhammad Izzudin Syamil - Computer Science Student University of Indonesia - GitHub
- Muhammad Ramadhan - Computer Science Student University of Indonesia - GitHub
- Github Language Used : Python 3.5.2 (https://www.python.org/)