rvqa

The Relation-VQA dataset is released!

If you have any question, please do not hesitate to contact me.

Relation-VQA Dataset

The dataset can be downloaded from the links: [Link 1], [Link 2].

Or you can download the data files one by one:

The whole dataset

• vg_vqa-rel_map_result_0.30.json
• vg_vqa-rel_map_result_0.30.txt

The splitted dataset

• vqa_raw_train_0.30.json
• vqa_raw_val_0.30.json
• vqa_raw_train_val_0.30.json
• vqa_raw_test_0.30.json

Alias rules

• alias_map_dict.json

Reference

• Paper on SIGKDD: http://www.kdd.org/kdd2018/accepted-papers/view/r-vqa-learning-visual-relation-facts-with-semantic-attention-for-visual-que
• Paper on arXiv: https://arxiv.org/abs/1805.09701

If you use this project as part of any published research, please acknowledge the following paper.

@inproceedings{lu2018rvqa,
	title={R-VQA: Learning Visual Relation Facts with Semantic Attention for Visual Question Answering.},
	author={Lu, Pan and Ji, Lei and Zhang, Wei and Duan, Nan and Zhou, Ming and Wang, Jianyong},
	booktitle={SIGKDD 2018},
	year={2018}
}