Video Faceswap

Action ID: video_faceswap

Description

Swap faces in videos using AI-powered deepfake technology. This node replaces faces in a target video with a face from a source image, maintaining natural expressions and movements throughout the video.

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_url	string	✓	-	URL of the source image containing the face to swap into the video
video_input_url	string	✓	-	URL of the input video where face swapping will be applied

View JSON Schema

{
  "description": "Video faceswap node input.",
  "properties": {
    "source_image_url": {
      "description": "Source image URL",
      "title": "Source image URL",
      "type": "string"
    },
    "video_input_url": {
      "description": "Video input URL",
      "title": "Video input URL",
      "type": "string"
    }
  },
  "required": [
    "source_image_url",
    "video_input_url"
  ],
  "title": "VideoFaceswapNodeInput",
  "type": "object"
}

Output Parameters

Name	Type	Description
video_url	string	URL of the processed video with face swap applied

View JSON Schema

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

How It Works

This node uses advanced AI deep learning models through PixelML to detect and replace faces in video content. The system analyzes each frame of the input video to identify facial features, expressions, and movements. It then seamlessly maps the face from the source image onto the detected faces in the video, preserving natural expressions, head movements, and lighting conditions. The processed video maintains the original video quality while applying the face swap throughout all frames.

Usage Examples

Example 1: Movie Scene Recreation

Input:

source_image_url: "https://example.com/images/actor-headshot.jpg"
video_input_url: "https://example.com/videos/movie-scene.mp4"

Output:

video_url: "https://pixelml.com/output/faceswap-abc123.mp4"

(Returns video with actor's face swapped into the movie scene)

Example 2: Marketing Campaign Personalization

Input:

source_image_url: "https://example.com/images/customer-photo.jpg"
video_input_url: "https://example.com/videos/product-demo.mp4"

Output:

video_url: "https://pixelml.com/output/faceswap-def456.mp4"

(Returns personalized video with customer's face in the product demonstration)

Example 3: Training Video Localization

Input:

source_image_url: "https://example.com/images/regional-trainer.jpg"
video_input_url: "https://example.com/videos/corporate-training.mp4"

Output:

video_url: "https://pixelml.com/output/faceswap-ghi789.mp4"

(Returns training video with regional trainer's face for localized content)

Common Use Cases

• Content Personalization: Create personalized video messages or advertisements featuring customer faces

• Film and Video Production: Generate preview shots or test scenes with different actors before final production

• Educational Content: Localize training videos by swapping instructors to represent different regions or cultures

• Social Media Content: Create entertaining videos with face swaps for viral marketing campaigns

• Virtual Try-On Experiences: Show customers how they would appear in promotional videos or advertisements

• Historical Recreation: Place modern faces into historical footage for educational or documentary purposes

• Gaming and Entertainment: Create custom character videos for games or interactive entertainment experiences

Error Handling

Error Type	Cause	Solution
No Face Detected in Source	Source image doesn't contain a clear face	Use a high-quality portrait image with a clearly visible face
No Face Detected in Video	Video doesn't contain detectable faces	Ensure the video has visible faces and good lighting throughout
Invalid Video Format	Video format is not supported	Convert video to a common format like MP4, AVI, or MOV
Video Too Long	Video exceeds maximum duration limit	Split the video into shorter segments or reduce the video length
Invalid Image URL	Source image URL is inaccessible or malformed	Verify the image URL is publicly accessible and correctly formatted
API Connection Failed	PixelML connection is invalid or expired	Check your PixelML credentials and connection status
Processing Timeout	Video processing took too long to complete	Try with a shorter video or lower resolution, then retry the request

Notes

• Source Image Quality: Use high-resolution, front-facing portrait images with clear facial features for best face swap results. Images should be well-lit and in focus.

• Video Quality: Higher quality input videos produce better results. Ensure the video has good lighting and clear facial visibility throughout.

• Face Orientation: Both the source image and video faces should be front-facing or similar angles. Extreme profile shots may produce less realistic results.

• Video Length: Processing time increases with video length. Shorter videos (under 30 seconds) process faster and more reliably.

• Multiple Faces: If the video contains multiple faces, the algorithm will typically swap all detected faces. For selective swapping, pre-process the video.

• Lighting Consistency: The AI attempts to match lighting conditions between source and target, but extreme lighting differences may affect realism.

• Resolution Preservation: The output video maintains the original video resolution, though some quality loss may occur during processing.

• Ethical Usage: Always obtain proper consent and use face swapping technology responsibly. Avoid creating misleading or harmful content.