> For the complete documentation index, see [llms.txt](https://facegsm.gitbook.io/facegsm-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://facegsm.gitbook.io/facegsm-documentation/facegsm/using-your-images.md).

# Using Your Images

## Description

User can use their **own custom face datasets** in FaceGSM, including personal photos, those of friends and family, or even targeted individuals. These photos are then collected in a **folder**, and FaceGSM will generate a new database file containing the embedding vectors for each face.

<figure><img src="/files/AMAUSxt6HDILjX25ruHr" alt=""><figcaption><p>Database Generator Program Flow</p></figcaption></figure>

*Note: By default, FaceGSM will use 100 Faces from VGGFace2.*

## Preparing Your Target Photos

1. Prepare your **own face datasets** in the required file formats (**jpg, jpeg,** or **png**) and store them in a folder
2. Run `facegsm.py database --dataset /path/to/dataset/folder` and wait for the process to finish
3. The new **JSON file** will be saved with the folder's name as its **filename** (`<folder's_name>.json`) in the `databases` folder
4. Update`DATABASE_PATH` in the `.env` using the newly generated JSON database file

## Example

```bash
python3 facegsm.py database --dataset /path/to/dataset/folder
```