#### News:
#### 2021.5.22: ---->RSITMD is expected to be released before July<----
#### 2021.6.21: ---->RSITMD is now open to access<----
#### 2021.10.28: ---->Four samples were updated to correct blank sentences<----
The similarity visualization results of six datasets, where the similarity score is weighted by the BLEU and METEOR indicators in the natural language processing field. The ideal picture is a straight diagonal line from the upper left to the lower right, which means each sentence is only related to the corresponding image.
Quantitative comparison of the four datasets. (a) Comparison of sample number. (b) Comparison of average sentence length. (c) Comparison of diversity score. (d) Comparison of average similarity. (e) Comparison of the total number of words. (f) Comparison of the number of categories.
Baidu Netdisk (Password:NIST) / Google Driver
File Structure:
## RSITMD:
## ---images # images
## ---dataset_RSITMD.json # annotations
## ---rsitmd_api.py # annotation parsing
If you feel this code helpful or use this code or dataset, please cite it as
Z. Yuan et al., "Exploring a Fine-Grained Multiscale Method for Cross-Modal Remote Sensing Image Retrieval," in IEEE Transactions on Geoscience and Remote Sensing, doi: 10.1109/TGRS.2021.3078451.