Visual QA (VQA)
Visual Question AnsweringEnglish
Visual QA (VQA) is a visual question answering dataset in English from Antol et al. with 265,016 images records in JSON format.
About Visual QA (VQA)
Dataset containing open-ended questions about images. These questions require an understanding of vision, language and commonsense to answer.
Details
- Task
- Visual Question Answering
- Language
- English
- Format
- JSON
- Rows / instances
- 265,016 images
- Creator
- Antol et al.
- Year
- 2015