Download
Download the cleaned data and extract them into "data" folder
- 1) http://tlberg.cs.unc.edu/licheng/referit/data/refclef.zip
- 2) http://tlberg.cs.unc.edu/licheng/referit/data/refcoco.zip
- 3) http://tlberg.cs.unc.edu/licheng/referit/data/refcoco+.zip
- 4) http://tlberg.cs.unc.edu/licheng/referit/data/refcocog.zip
Prepare Images:
Besides we add "mscoco" into the "data/images" folder. Download it from mscoco This dataset is for refcoco, refcoco+ and refgoogle. For refclef, we add "saiapr_tc-12" into 'data/images' folder. I only extracted the related images as a subset of the original imageCLEF, i.e., 19997 images. Please download the subset from here (http://tlberg.cs.unc.edu/licheng/referit/data/images/saiapr_tc-12.zip). Our dataset API "pyutils/datasets/refer.py" is able to load all 4 datasets.
How to use
# locate your own data_root, and choose the dataset_splitBy you want to use
refer = REFER(data_root, dataset='refclef', splitBy='unc')
refer = REFER(data_root, dataset='refcoco', splitBy='unc')
refer = REFER(data_root, dataset='refcoco', splitBy='google')
refer = REFER(data_root, dataset='refcoco+', splitBy='unc')
refer = REFER(data_root, dataset='refcocog', splitBy='google')