Textual Visual Semantic Dataset
Automatic Image CaptioningEnglish
Textual Visual Semantic Dataset is a automatic image captioning-focused dataset in English that provides 82 labeled examples distributed in JPG, CSV format.
About Textual Visual Semantic Dataset
A dataset consisting of detecting and recognizing text appearing in images (e.g. signboards, traffic signals or brands in clothing or objects). Around 82,000 images.
Details
- Task
- Automatic Image Captioning
- Language
- English
- Format
- JPG, CSV
- Rows / instances
- 82
- Creator
- Sabir et al.
- Year
- 2020