#### Core_Dataset: 

- 69 categories, 
- 582 images, 
- 582 text annotations.

#### Style_Dataset: 

- 5 categories, 
- 37 images, 
- 37 text annotations.

#### Vedio_Dataset:

- 35 categories, 
- 72 videos.