Reference

(1) Code : https://github.com/sgrvinod/a-PyTorch-Tutorial-to-Image-Captioning
(2) Paper : https://arxiv.org/abs/1502.03044

Data setup

Data

wget http://images.cocodataset.org/zips/train2014.zip
wget http://images.cocodataset.org/zips/val2014.zip
wget http://images.cocodataset.org/zips/test2014.zip

Annotation

Download splits and captions in this link

Quick implementation

python create_input_files.py
python train.py
python eval.py
python caption.py --img='path/to/img.jpg' \
--model='path/to/model.pth.tar' \
--word_map='path/to/WORDMAP.json' --beam_size=5