This tutorial only applies to UNIX like systems. If your system not an UNIX type, please search the tutorial based on your system.
- Miniconda with Python 3
- Docker
wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh
conda env create
source activate image-captioning
mkdir dataset
cd dataset
M. Hodosh, P. Young and J. Hockenmaier (2013) "Framing Image Description as a Ranking Task: Data, Models and Evaluation Metrics", Journal of Artifical Intellegence Research, Volume 47, pages 853-899 http://www.jair.org/papers/paper3994.html
wget http://nlp.cs.illinois.edu/HockenmaierGroup/Framing_Image_Description/Flickr8k_Dataset.zip
wget http://nlp.cs.illinois.edu/HockenmaierGroup/Framing_Image_Description/Flickr8k_text.zip
unzip Flickr8k_Dataset.zip
unzip Flickr8k_text.zip
python -m src.train -m <your-model-type>
1: train using image caption sentence modeler
2: train using image caption single word modeler
python -m src.train -m <your-model-type>
1: serving using image caption sentence modeler
2: serving using image caption single word modeler
python -m src.predict -m <your-model-type> -p <your-image-path> [OPTION]
1: predict using image caption sentece modeler
2: predict using image caption single word modeler
- --show-image : show captioned image with matplotlib
python -m src.run_video <your-video-path>
If video path is empty, it will use your webcam as video stream
python -m src.run_cctv
- Aditya Alif Nugraha - Aditya Alif Nugraha
- Bramantyo Ardian