ilkerkesen / colorfromlanguage

Code base of the paper : Learning to Color from Language

Learning to Color from Language

Learning to Color from Language, NAACL, 2018 (short)

Instructions :

1. Download data : https://obj.umiacs.umd.edu/learning_to_color/coco_colors.h5

2. Extract features from black-and-white images python extract_image_features.py --input_h5_file coco_colors.h5 --output_h5_file image_features.h5

3. Then run : python autocolorize_resnet.py --h5_File coco_colors.h5 --features_file ./image_features.h5 --vocab_file_name ./priors/coco_colors_vocab.p --image_save_folder ./trial/ --model_save_file ./models/

Film Activations

If you use this work, please cite:

@InProceedings{Manjunatha:Iyyer-ColorLanguage,
    Title = {Learning to Color from Language},
    Booktitle = {North American Chapter of the Association for Computational Linguistics},
    Author = {Varun Manjunatha and Mohit Iyyer and Jordan Boyd-Graber and Larry Davis},
    Year = {2018},
}