VQA Benchmark Task
A VQA Benchmark Task is a QA benchmark task for VQA tasks.
- See: Vision Task, Language Task, Commonsense Task.
References
2015
- http://visualqa.org/
- QUOTE: VQA is a new dataset containing open-ended questions about images. These questions require an understanding of vision, language and commonsense knowledge to answer.
- 254,721 images (MSCOCO and abstract scenes)
- 3 questions per image (764,163 total)
- 10 ground truth answers per question
- 3 plausible (but likely incorrect) answers per question
- Open-ended and multiple-choice answering tasks.
- Automatic evaluation metric.
- QUOTE: VQA is a new dataset containing open-ended questions about images. These questions require an understanding of vision, language and commonsense knowledge to answer.