martin-ev/vizwiz_vgoq

---
license: cc-by-nc-4.0
source_datasets:
- VizWiz-Visual-Question-Answering
- VizWiz-Answer-Grounding-for-VQA
---

This synthetic dataset is built on top of the [VizWiz-Visual-Question-Answering dataset](https://vizwiz.org/tasks-and-datasets/vqa/) and on the [VizWiz-Answer-Grounding-for-VQA dataset](https://vizwiz.org/tasks-and-datasets/answer-grounding-for-vqa/).
The first contains images taken by people who are blind or visually impaired using mobile devices, as well as visual questions and answers.
The second contains grounding annotation, locating directly visible answers to the question.

The fields `object_question` and `grounding_evidential_quality` contain our new synthetically generated questions/annotations.

**License (source datasets):** The two source datasets (VizWiz-Visual-Question-Answering dataset and VizWiz-Answer-Grounding-for-VQA dataset) were originally released under the [Creative Commons Attribution 4.0 International License](https://creativecommons.org/licenses/by/4.0/).

**License (our synthetic questions/annotations):** Our newly generated questions/annotations are licensed under the [Creative Commons Attribution-NonCommercial 4.0 International License](https://creativecommons.org/licenses/by-nc/4.0/). Because the new questions/annotations are generated using Claude 3.7 Sonnet, [Anthropic's usage policies](https://www.anthropic.com/legal/aup) also apply, see especially [model training with Claude's outputs](https://support.claude.com/en/articles/12326764-can-i-use-my-outputs-to-train-an-ai-model).