> For the complete documentation index, see [llms.txt](https://docs.agenticflow.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.agenticflow.ai/reference/nodes/video_faceswap_pro.md).

# Video Faceswap Pro

**Action ID:** `video_faceswap_pro`

## Description

Advanced face swapping for videos with support for multiple faces and highest quality output

## Connection

| Name               | Description                                 | Required | Category |
| ------------------ | ------------------------------------------- | -------- | -------- |
| PixelML Connection | The PixelML connection to call PixelML API. | True     | pixelml  |

## Input Parameters

| Name                | Type   | Required | Default | Description                                                                                                     |
| ------------------- | ------ | :------: | ------- | --------------------------------------------------------------------------------------------------------------- |
| source\_image\_urls | array  |     ✓    | -       | Array of source image URLs containing faces to swap into the video                                              |
| video\_input\_url   | string |     ✓    | -       | URL of the input video where faces will be swapped                                                              |
| face\_detect\_info  | object |     ✓    | -       | Face detection information containing crop\_arr, crop\_landmarks, and face\_index data for precise face mapping |

<details>

<summary>View JSON Schema</summary>

**Input Schema**

```json
{
  "description": "Video faceswap pro node input.",
  "properties": {
    "source_image_urls": {
      "description": "Source image URLs",
      "items": {
        "type": "string"
      },
      "title": "Source image URLs",
      "type": "array"
    },
    "video_input_url": {
      "description": "Video input URL",
      "title": "Video input URL",
      "type": "string"
    },
    "face_detect_info": {
      "additionalProperties": true,
      "description": "Face detection information containing crop_arr, crop_landmarks, and face_index data",
      "title": "Face detect info",
      "type": "object"
    }
  },
  "required": [
    "source_image_urls",
    "video_input_url",
    "face_detect_info"
  ],
  "title": "VideoFaceswapProNodeInput",
  "type": "object"
}
```

</details>

## Output Parameters

| Name       | Type   | Description                                   |
| ---------- | ------ | --------------------------------------------- |
| video\_url | string | URL of the processed video with swapped faces |

<details>

<summary>View JSON Schema</summary>

**Output Schema**

```json
{
  "description": "Video faceswap pro node output.",
  "properties": {
    "video_url": {
      "description": "Video URL",
      "title": "Video URL",
      "type": "string"
    }
  },
  "required": [
    "video_url"
  ],
  "title": "VideoFaceswapProNodeOutput",
  "type": "object"
}
```

</details>

## How It Works

This advanced node uses state-of-the-art AI face swapping technology to replace faces in videos with the highest quality output. It requires face detection information (crop\_arr, crop\_landmarks, face\_index) to precisely map and swap faces. The node supports multiple face swapping in a single video, maintaining natural lighting, expressions, and movements while ensuring the swapped faces blend seamlessly with the original video.

## Usage Examples

### Example 1: Single Face Swap

**Input:**

```
source_image_urls: ["https://example.com/person-a.jpg"]
video_input_url: "https://example.com/original-video.mp4"
face_detect_info: {
  "crop_arr": [[100, 200, 300, 400]],
  "crop_landmarks": [[[150, 220], [250, 220], [200, 280], [150, 330], [250, 330]]],
  "face_index": [0]
}
```

**Output:**

```
video_url: "https://storage.pixelml.com/swapped-video.mp4"
```

### Example 2: Multiple Face Swap

**Input:**

```
source_image_urls: [
  "https://example.com/person-a.jpg",
  "https://example.com/person-b.jpg"
]
video_input_url: "https://example.com/group-video.mp4"
face_detect_info: {
  "crop_arr": [[100, 200, 300, 400], [500, 200, 700, 400]],
  "crop_landmarks": [
    [[150, 220], [250, 220], [200, 280], [150, 330], [250, 330]],
    [[550, 220], [650, 220], [600, 280], [550, 330], [650, 330]]
  ],
  "face_index": [0, 1]
}
```

**Output:**

```
video_url: "https://storage.pixelml.com/multi-swap-video.mp4"
```

### Example 3: High-Quality Production

**Input:**

```
source_image_urls: ["https://example.com/actor-headshot.jpg"]
video_input_url: "https://example.com/scene-footage.mp4"
face_detect_info: {
  "crop_arr": [[150, 180, 450, 520]],
  "crop_landmarks": [[[200, 250], [400, 250], [300, 350], [200, 430], [400, 430]]],
  "face_index": [0]
}
```

**Output:**

```
video_url: "https://storage.pixelml.com/production-video.mp4"
```

## Common Use Cases

* **Content Creation**: Create videos featuring different people without reshooting
* **Film Production**: Replace actors in scenes for visual effects or reshoots
* **Marketing Videos**: Personalize video content by swapping faces for different audiences
* **Educational Content**: Create training videos with different instructors without multiple recordings
* **Entertainment**: Create fun face swap videos for social media or personal use
* **Video Dubbing**: Combine with lip sync for complete character replacement in localized content
* **Privacy Protection**: Replace faces in videos while maintaining video quality and natural appearance

## Error Handling

| Error Type                  | Cause                                                | Solution                                                                                  |
| --------------------------- | ---------------------------------------------------- | ----------------------------------------------------------------------------------------- |
| Invalid Face Detection Data | face\_detect\_info is malformed or incomplete        | Ensure face\_detect\_info contains valid crop\_arr, crop\_landmarks, and face\_index data |
| Face Count Mismatch         | Number of source images doesn't match detected faces | Provide the same number of source images as faces to swap                                 |
| No Faces Detected           | Video doesn't contain detectable faces               | Use videos with clearly visible faces                                                     |
| Invalid Video Format        | Video format is not supported                        | Use common video formats like MP4, MOV, or AVI                                            |
| Image Quality Too Low       | Source images are too low resolution                 | Use high-quality source images for best results                                           |
| Video Too Long              | Video exceeds maximum duration limit                 | Split video into shorter segments                                                         |
| Connection Failed           | Unable to access PixelML API                         | Check PixelML connection credentials and API availability                                 |

## Notes

* **Face Detection Required**: This node requires face detection information. Use a face detection node first to generate the required face\_detect\_info data.
* **Image Quality**: Use high-resolution source images for best face swap quality. Clear, well-lit photos work best.
* **Video Quality**: The output quality depends on both the source images and input video quality.
* **Multiple Faces**: The Pro version supports swapping multiple faces simultaneously in a single video.
* **Processing Time**: Processing time varies based on video length and number of faces. Expect 1-5 minutes for typical videos.
* **Natural Results**: The node maintains facial expressions, head movements, and lighting conditions for realistic results.