Decoupled Novel Object Captioner

TensorFlow code for our paper [Link].

Preparation

Dependencies

• Python 3.6
• Tensorflow (version >= 1.4.1)
• nltk, tqdm

Prepare data

Please follow the instructions in prepare_data.

Train

python run.py --stage train

It takes about one hour to train the model with a Nvidia V100 GPU.

Test

python run.py --stage test

Our trained checkpoint can be found here.

Reference

Please cite the following paper in your publications if it helps your research:

@inproceedings{wu2018decoupled,
  title     = {Decoupled Novel Object Captioner},
  author    = {Wu, Yu and Zhu, Linchao and Jiang, Lu and Yang, Yi},
  booktitle = {Proceedings of the 2018 ACM on Multimedia Conference (ACM MM)},
  year      = {2018}
}

Contact

To report issues for this code, please open an issue on the issues tracker.

If you have further questions about this paper, please do not hesitate to contact me.

Yu Wu's Homepage