Code base for our paper " Adversarial Scene Editing: Automatic Object Removal from Weak Supervision" appearing in NIPS 2018.
- Upload dataset file and pre-trained model
- Usage instruction
- port to python 3.6
- port to pytorch 1.0
- Pre-trained model - Weakly supervised removal model for 20 pascal object categories trained on COCO dataset.
- COCO dataset file - Single json file with metadata and annotations for the COCO dataset
- Poster
If you find this code useful in your work, please cite the paper.
@inproceedings{shetty_neurips2018,
TITLE = {Adversarial Scene Editing: Automatic Object Removal from Weak Supervision},
AUTHOR = {Shetty, Rakshith and Fritz, Mario and Schiele, Bernt},
PUBLISHER = {Curran Associates},
YEAR = {2018},
BOOKTITLE = {Advances in Neural Information Processing Systems 31},
PAGES = {7716--7726},
ADDRESS = {Montr{\'e}al, Canada},
}
Lot of the code structure is borrowed from the Stargan repository (https://github.com/yunjey/stargan)