henghuiding / ReLA

[CVPR2023 Highlight] GRES: Generalized Referring Expression Segmentation
https://henghuiding.github.io/GRES/
MIT License
656 stars 19 forks source link

Can you provide the dataset preparation and the download links? #1

Closed yahooo-m closed 1 year ago

henghuiding commented 1 year ago

Thank you for your attention. For dataset download, please refer to here, we will update the document for more detailed preparation within one week.

yahooo-m commented 1 year ago

Thanks! This is really a great job!

yahooo-m commented 1 year ago

Could you please show me the directory structure of the datasets folder?

changliu19 commented 1 year ago

Could you please show me the directory structure of the datasets folder?

A basic structure of the folder datasets would be like this:

datasets
├── grefcoco
│   ├── grefs(unc).json
│   ├── instances.json
├── images
│   └── train2014
├── refcoco
│   ├── instances.json
│   ├── refs(google).p
│   └── refs(unc).p
├── refcoco+
│   ├── instances.json
│   └── refs(unc).p
└── refcocog
    ├── instances.json
    ├── refs(google).p
    └── refs(umd).p

We will be updating the detailed documentation and making the code easier to use in the next few days. Please stay tuned!

yahooo-m commented 1 year ago

Thanks! The code is clear.

Seonghoon-Yu commented 1 year ago

Thanks for releasing the code and pre-trained weights.

changliu19 commented 1 year ago

Documents and code are updated.