This repo contains pre-trained model checkpoints ONLY. The main repo is here.

Released on 05 June 2019.

Just point infer.py to the directory containing the checkpoints. Model configurations are loaded from config.pkl.

For example:

python infer.py   \
    --infer_checkpoints_dir mscoco/radix_b256_add_LN_softmax_h8_tie_lstm_cnnFT_run_01

The checkpoints should be compatible with release v1.0 through v1.1.

The size of the checkpoints may be larger as they include CNN weights.

If you find this repository useful for your research or work, please cite:

@article{tan2019comic,
  title={COMIC: Towards A Compact Image Captioning Model with Attention},
  author={Tan, Jia Huei and Chan, Chee Seng and Chuah, Joon Huang},
  journal={IEEE Transactions on Multimedia},
  year={in Press},
  publisher={IEEE}
}

Suggestions and opinions (both positive and negative) are greatly welcomed. Please contact the authors by sending an email to tan.jia.huei at gmail.com or cs.chan at um.edu.my.

The project is open source under BSD-3 license (see the LICENSE file).

© 2019 Center of Image and Signal Processing, Faculty of Computer Science and Information Technology, University of Malaya.