201528014227051/rsicd_optimal

Datasets for remote sensing images (Paper:Exploring Models and Data for Remote Sensing Image Caption Generation)

RSICD

❗ News

• The new download source of RSICD-MEGA

• The new download source of Sydney-captions and UCM-catpions-MEGA.

Intruduction

RSICD is used for remote sensing image captioning task. more than ten thousands remote sensing images are collected from Google Earth, Baidu Map, MapABC, Tianditu. The images are fixed to 224X224 pixels with various resolutions. The total number of remote sensing images are 10921, with five sentences descriptions per image. To the best of our knowledge, this dataset is the largest dataset for remote sensing captioning. The sample images in the dataset are with high intra-class diversity and low inter-class dissimilarity. Thus, this dataset provides the researchers a data resource to advance the task of remote sensing captioning.

Git downloads are size limited, need to use large file storage (LFS), and also need to modify the settings, specifically see here. It is recommended that you download from the Google SkyDrive and Baidu SkyDrive.

注意：git下载是有大小限制的，需要使用大文件存储（LFS）,还需要修改设置，具体参见这里。推荐大家使用下面的谷歌网盘和百度网盘下载。

Examples

There are two examples in dataset:

Citation

If you find this dataset useful, please cite this paper：

	@article{lu2017exploring,
	 title={Exploring Models and Data for Remote Sensing Image Caption Generation},
	 author={Lu, Xiaoqiang and Wang, Binqiang and Zheng, Xiangtao and Li, Xuelong},
	 journal={IEEE Transactions on Geoscience and Remote Sensing},
	 volume = {56},
	 number = {4},
	 pages = {2183-2195},
	 doi={10.1109/TGRS.2017.2776321}
	}

Other Source of RSICD Datasets

Baidupan

GoogleDrive

Other remote sensing image captioning dataset

UCM_captions-BaiduPan

Sydney_captions-BaiduPan

UCM_captions-MEGA

RSICD-MEGA

Sydney_captions-MEGA

rsicd_optimal's Issues

UCM_captions does not contain all the images mentioned in dataset.json

I see that the dataset.json from UCM_captions has captions for 2100 images, but the imgs.rar only contains 613 images. Where can the full dataset be found?

source code

Hello, I am interested in this excellent work. Can you provide the source code for studying?

do you have captionval json file?

I was going to do some evluation on your dataset but i find there wasn't captionval json file,can you provide or tell me how to create it?

Sydney Caption Dataset

Please provide sydney caption dataset. I cannot access the Baidu link. It's really needed for my thesis work. I can digitally sign on documents indicating it will not be used for commercial purpose. My email id: [email protected]. Thanks in advance.

Location of overhead imagery in the dataset

Hi,

Do you know the geospatial locations of the imagery in the RSICD dataset? Ideally, precise lat/lons but country or city level will also be helpful.

Thanks

Datasets of UCM and Sydney with google drive instead of Baidu

Could you provide the datasets of UCM and Sydney with google drive? Right now to download the datasets, it is required: “The content you download contains folder, please use the Baidu network disk client to download”.

Sydney_captions- MEGA No file available

There is no file available on the link of sydney_captions. Please add the releated files to the link.

Thanks in advance.

Too many typos

Could you please update the captions? Too many typos there. Afraid that I have to give up this dataset.

RSID Data label name

Has anyone noticed that the name of the tagged image in the RSID dataset does not correspond to the name of the image in the image? When I checked this dataset and wanted to use it, I found that the name of the image in json was inconsistent with the name of the original image. Could you please tell me how to correct this problem

Can you provide the original url of these dataset?

Hi!

I have read your paper. It is a good work! Thanks for your sharing the datasets. But I have a few questions. Do these datasets are open datasets? Can you provide the original url of these datasets?

I can only find the UCM datasets. "http://weegee.vision.ucmerced.edu/datasets/landuse.html".

Thanks a lot! :)