---
license: cc-by-sa-4.0
---

# syn-vis-v0-base

Mirror of the exact bmcore v24 holdout subset. Source: [retowyss/Syn-Vis-v0](https://huggingface.co/datasets/retowyss/Syn-Vis-v0).

All 480 local basenames match images/base/. Card licenses images CC0 and compilation/documentation CC BY-SA 4.0.

Contains 480 synthetic image files. This mirror repackages the media; it does not grant additional rights.

Source revision reviewed: `100262732989e77f38cd831d70a376a93735006a`.

## Original source card

---
license: cc-by-sa-4.0
language:
- en
tags:
- synthetic
- faces
- portraits
- face-generation
- semi-realistic
pretty_name: Syn-Vis-v0
size_categories:
- n<1K
---

# Syn-Vis-v0: A Dataset of Synthetic Faces

Syn-Vis-v0 (Synthetic Visage Version 0) is a dataset of 480 synthetic faces generated with Qwen-Image and Qwen-Image-Edit-2509.

![Face Samples](misc/banner-md.png)

- **Diversity**:
  - The dataset is balanced across ethnicities - approximately 60 images per broad category (Asian, Black, Hispanic, White, Indian, Middle Eastern) and 120 ethnically ambiguous images.
  - Wide range of skin-tones, facial features, hairstyles, hair colors, nose shapes, eye shapes, and eye colors.
- **Quality**:
  - Rendered at 2048x2048 resolution using Qwen-Image-Edit-2509 (BF16) and 50 steps.
  - Checked for artifacts, defects, and watermarks.
- **Style**: semi-realistic, 3d-rendered CGI, with hints of photography and painterly accents.
- **Captions**: Natural language descriptions consolidated from multiple caption sources using GPT-OSS-120B.
- **Metadata**: Each image is accompanied by ethnicity/race analysis scores (0-100) across six categories (Asian, Indian, Black, White, Middle Eastern, Latino Hispanic) generated using DeepFace.
- **Analysis Cards**: Each image has a corresponding analysis card showing similarity to other faces in the dataset.

## Updates

- **2025-11-01**: Added a second set of images. These images have a higher shot-type consistency (head-and-shoulders portraits).
  - Directory: `images/headshot/`
  - Manifest: `metadata_headshot.csv`
  - Resolution: 1024×1024 (50 steps, CFG 4, Qwen-Image-Edit-2509 BF16)
  - Metadata: Only filename links to corresponding base image; demographic scores are not re-computed.
- **2025-10-24**: Improved dataset structure to better conforms with `datasets` library standards. Added `dataset_info.json` for schema definition. Updated README with more detailed information.

## Analysis Cards

![Face Similarities](misc/embedding-distances.png)

![Analysis Card (001-0051)](misc/cards/001-0051.png)

## Dataset Structure

```txt
Syn-Vis-v0/
├── images/
│   └── base/             # Main dataset images
│   └── headshot/         # Additional headshot-style images
├── metadata.csv          # Root-level metadata file for Hugging Face preview
├── metadata_headshot.csv
├── dataset_info.json     # Schema definition for image and metadata fields
├── misc/                 # Analysis cards, banners, etc.
└── README.md
```

## Metadata Format

The `metadata.csv` contains the following columns:

- `file_name`: Image filename (e.g., "001-0042.png")
- `caption`: Consolidated natural language description
- `race_asian`: Asian demographic score (0-100)
- `race_indian`: Indian demographic score (0-100)
- `race_black`: Black demographic score (0-100)
- `race_white`: White demographic score (0-100)
- `race_middle_eastern`: Middle Eastern demographic score (0-100)
- `race_latino_hispanic`: Latino Hispanic demographic score (0-100)
- `dominant_race`: Primary predicted demographic category

## Caption Quality

Captions are consolidated from multiple JoyCaption-Beta-One outputs using GPT-OSS-120B and feature:

- **Natural language**: Start with "The woman..." or "A woman..."
- **Specific descriptions**: Physical features described precisely rather than using broad demographic categories
- **Structured order**: Face → hair/physical features → clothing → background → lighting
- **Style-neutral**: Remove technical photography terms and medium references
- **Flowing narrative**: Natural sentences without section headers

Example caption:
> "The woman has a smooth medium‑brown complexion that catches a gentle, even glow. Her eyes are large, dark brown and framed by thick, dark lashes, giving them a calm, slightly serious look as she gazes directly forward..."

## Use Cases

- Raw training data for small models
- Base images for image-to-image generation tasks
- Base images for style transfer
- Whatever you want!

## Statistics

- **Ethnicities**: (by Dominant Race Counts; some faces show similar scores across multiple categories)
  - White: 94 images
  - Latino Hispanic: 93 images
  - Asian: 90 images
  - Indian: 70 images
  - Black: 68 images
  - Middle Eastern: 65 images
- **Skin tones**: Full spectrum from very light to very dark
- **Facial features**: Wide variety of eye shapes, nose shapes, lip shapes
- **Hair styles**: Various textures, colors, and arrangements
- **Backgrounds**: Dark and light, plain and scenic
- **Ages**: Almost exclusively 30 ± 5 years according to analysis with DeepFace.

> Anecdotally, Asian, White, and Black, were predicted with a single high score (85+) much more frequently than Latino Hispanic, Indian and Middle Eastern.

## Ethical Considerations and Other Notes

- The ethnicity/race labels are generated by automated analysis and should not be considered ground truth for real-world applications involving human subjects. Their primary purpose is to ensure coverage of wide range facial features.
- Only female-presenting individuals are included. I decided against including male-presenting individuals because **beards** - I didn't know how well the classifiers would handle them (obscured features), so I decided to avoid that complexity.
- All faces were explicitly declared female-presenting (in the prompt and the caption), however, DeepFace occasionally suggested some images may be male-presenting.
- The dataset has a strong beauty bias and the faces are unusually symmetrical.

## Creation Process

1. **Initial Image Generation**: Generated an initial set of 5,500 images at 768x768 using Qwen-Image (FP8). Facial features were randomly selected from lists and then written into natural prompts by Qwen3:30b-a3b. The style prompt was "Photo taken with telephoto lens (130mm), low ISO, high shutter speed".
2. **Initial Analysis & Captioning**: Each of the 5,500 images was captioned three times using JoyCaption-Beta-One. These initial captions were then consolidated using Qwen3:30b-a3b. Concurrently, demographic analysis was run using DeepFace.
3. **Selection**: A balanced subset of 480 images was selected based on the aggregated demographic scores and visual inspection.
4. **Enhancement**: Minor errors like faint watermarks and artifacts were manually corrected using GIMP.
5. **Upscaling & Refinement**: The selected images were upscaled to 2048x2048 using Qwen-Image-Edit-2509 (BF16) with 50 steps at a CFG of 4. The prompt guided the model to transform the style to a high-quality 3d-rendered CGI portrait while maintaining the original likeness and composition.
6. **Final Captioning**: To ensure captions accurately reflected the final, upscaled images and accounted for any minor perspective shifts, the 480 images were fully re-captioned. Each image was captioned three times with JoyCaption-Beta-One, and these were consolidated into a final, high-quality description using GPT-OSS-120B.
7. **Final Analysis**: Each final image was analyzed using DeepFace to generate the demographic scores and similarity analysis cards present in the dataset.

### Models and Tools Used

- **Qwen-Image**: Image Generation
- **Qwen-Image-Edit-2509**: Image Refinement/Upscaling
- **JoyCaption-Beta-One**: Captioning
- **Qwen3:30b-a3b**: Prompt Writing & Initial Caption Consolidation
- **GPT-OSS-120B**: Final Caption Consolidation
- **Tools**: vLLM, DeepFace, Python, R, GIMP, ComfyUI

## Projects That Use Syn-Vis-v0

- Coming soon!
- Your project here?

## Known Issues

- 001-0309: Appears to be wearing a mask, likely introduced during the image-to-image upscaling step.

## License

This project is released under a dual-license structure:

- **Images**: All individual images in this dataset are dedicated to the public domain via the **Creative Commons CC0 1.0 Universal Public Domain Dedication**. You are free to copy, modify, distribute, and perform the work, even for commercial purposes, all without asking permission.
- **Dataset & Documentation**: The specific compilation of this dataset, its metadata, and all associated documentation (the "Curation") are licensed under the **Creative Commons Attribution-ShareAlike 4.0 International License (CC-BY-SA 4.0)**.

Interpretation of the License:

- You can use the images for any project (commercial or non-commercial) with no attribution and no restrictions.
- You can use and share the entire dataset/documentation, but you must give credit to us (BY) and share any modifications to the curation itself under the same (SA) license.

## Citation

```bibtex
@dataset{syn-vis-v0-2025,
  title={Syn-Vis-v0: A Synthetic Face Dataset},
  author={Wyss, Reto},
  year={2025},
  url={https://huggingface.co/datasets/retowyss/Syn-Vis-v0},
  note={Images: CC0 (Public Domain); Dataset compilation and documentation: CC-BY-SA-4.0}
}
```