Merge Portraits API

Request

Image requirements

Field	Requirements
image_target	Image format: JPEG JPG BMP PNG Image size: No more than 4 MB. Image resolution: Larger than 128x128px, smaller than 4096x4096px. Face pixel size: To ensure the fusion effect, it is recommended that the minimum value of the side length of the face box (square) in the image is not less than 200px. Face quality: The higher the face quality, the better the fusion effect. Factors affecting face quality include: occlusion of the five facial features, improper lighting (bright light, dark light, backlighting), excessive face angle (recommended yaw ≤ ±20°, pitch ≤ ±20°), etc. Black and white images are not supported.
image_template	Image format: JPEG JPG BMP PNG Image size: No more than 4 MB. Image resolution: Larger than 200x200px, smaller than 4096x4096px. Note that for special face materials, such as cartoon style images with large eyes, the original key point results will be deviated, and should be made accurate by dragging the position in the configuration tool. Most normal images are already very accurate and do not need to be adjusted. The pixel area of the face in the image should not be too small (at least 200x200px, too small to change the face will not be clear), nor too large (the pixel size of the face area and speed is positively correlated, too large will affect the server speed and increase costs). Pay attention to the quality of the material, make sure the face is clear enough, there should be no noise caused by compression, otherwise it will reduce the quality of the face replacement result. For better results, the face of the material should be as positive as possible, with the highest yaw angle required (within plus or minus 10 degrees recommended), followed by the pitch angle (within plus or minus 20 degrees recommended), and the roll angle (within plus or minus 30 degrees).

Body

Field	Required	Type	Scope	Description
image_target	YES	file		Target image.
image_template	YES	file		Template images.
source_similarity	NO	float	[0, 1]	Face similarity, where a higher numerical value indicates greater similarity. 0: Consistent with the original template. 1: Maximum similarity with the target image.

Response

Response Field Handling Flow

1. Handle Public Response Fields

   Parse and validate the Public Response Fields, checking the status code or response message to ensure the request is successful and error-free.

2. Handle Business Response Fields

   If the Public Response Fields are valid and error-free, proceed with processing the business logic in the Business Response Fields.

Public Response Fields

Viewing Public Response Fields and Error Codes

Business Response Fields

Field	Type	Description
data	object	The content of the result data returned.
+image	string	The result image, returning the Base64 encoding of the image.

Response Example

{
  "request_id":     "",
  "log_id":         "",
  "error_code":     0,
  "error_msg":      "",
  "error_detail":   {
    "status_code":  200,
    "code":         "",
    "code_message": "",
    "message":      ""
  },
  "data":           {
    "image": ""
  }
}