---
license: cc-by-4.0
task_categories:
  - image-to-text
  - object-detection
  - visual-question-answering
language:
  - en
size_categories:
  - 100K<n<1M
tags:
  - visual-genome
  - scene-graph
  - region-descriptions
  - visual-relationships
  - attributes
  - object-detection
  - image-captioning
  - vqa
pretty_name: Visual Genome Raw Data
---

# Visual Genome Raw Data

This repository provides a convenient copy of the raw Visual Genome dataset files for easier access and downloading.

> This is **not an official Visual Genome release**.  
> The original dataset was created and released by the Visual Genome authors.  
> Please refer to the official Visual Genome website and paper for authoritative information.

## Dataset Summary

Visual Genome is a large-scale vision-language dataset with dense image annotations, including:

- Images
- Region descriptions
- Object annotations
- Attribute annotations
- Relationship annotations
- Scene graphs
- Visual question-answering annotations
- WordNet synset mappings

The dataset contains approximately:

- 108K images
- 5.4M region descriptions
- 1.7M visual question-answer pairs
- 3.8M object instances
- 2.8M attributes
- 2.3M relationships

## Intended Use

This dataset can be used for research on:

- Image-text representation learning
- Image captioning
- Dense captioning
- Object detection
- Attribute recognition
- Visual relationship detection
- Scene graph generation
- Visual question answering
- Vision-language pretraining

## Source

Original Visual Genome website:

```text
https://visualgenome.org/
```

Dataset page:

```text
https://homes.cs.washington.edu/~ranjay/visualgenome/
```

## License

Visual Genome is licensed under the Creative Commons Attribution 4.0 International License:

```text
CC BY 4.0
https://creativecommons.org/licenses/by/4.0/
```

Users should comply with the original license terms, including proper attribution to the Visual Genome authors.

## Citation

If you use this dataset, please cite the original Visual Genome paper:

```bibtex
@article{krishna2017visualgenome,
  title={Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations},
  author={Krishna, Ranjay and Zhu, Yuke and Groth, Oliver and Johnson, Justin and Hata, Kenji and Kravitz, Joshua and Chen, Stephanie and Kalantidis, Yannis and Li, Li-Jia and Shamma, David A and Bernstein, Michael and Fei-Fei, Li},
  journal={International Journal of Computer Vision},
  year={2017}
}
```

## Disclaimer

This repository is not affiliated with, endorsed by, or maintained by the original Visual Genome authors.

The data is provided as-is for convenience. Users are responsible for verifying the dataset version, file integrity, license terms, and suitability for their own use cases.