Compare the photo and the video uploaded through the interface to determine whether the faces belong to the same person.
For photo and video standards, refer to Chapter 3 Photo and Video Standards in the Operation Manual.
POST
check_quality
can currently only evaluate face occlusion. Other evaluation items may be added in the future.
The possible results ofliveness_status
are as follows:
Correlations between the verification score threshold and the error rate:
Recommended threshold: greater than 0.7.
Name
Type
Required
Description
encrypted_video
string
Yes
Encrypted video Encryption method reference: [Photo and Video Encryption]
encrypted_image
string
Yes
Encrypted photo Encryption method reference: [Photo and Video Encryption]
auto_rotate
boolean
No
The default value is false, indicating that the photo is not rotated. When the value is true, the photo is automatically rotated.
check_quality
boolean
No
The default value is false, and no quality test is performed. When the value is true, the photo quality is checked.
return_image
boolean
No
Whether to return the selected frame of face and timestamp of the selected frame of the photo. The default is false. image_timestamp
and base64_image
are returned only when this value is true and the live detection is passed.
return_face_image
boolean
No
Whether to return the face clipping of the selected frame of the photo. The default is false. base64_face_image
is returned only when this value is true and the live detection is passed.
return_status
boolean
No
Whether to return the error status description. The default is false. liveness_status
is returned only when this value is true.
Name
Type
Description
code
int
System response code: 1000
passed
boolean
Liveness detection passed or not
liveness_score
float
The score of silent liveness detection (for reference only, please refer to the passed field)
liveness_status
string
Description of the error status of silent liveness detection. This field is returned when return_status=true
verification_score
float
Face comparison score, recommended threshold: > 0.7. This field is returned when passed=true
image_timestamp
float
Timestamp of the selected frame of image in seconds. This field is returned only when passed=true and return_image=true
base64_image
string
The selected frame of the photo. This field is returned only when passed=true and return_image=true
base64_face_image
string
Face clipping of the selected frame of the photo. This field is returned only when passed=true and return_face_image=true
request_id
string
The ID of this request
Status
Description
ok
Silent liveness detection passed: the person in the video is real
hack
Silent liveness detection failed, reason: face spoofing (for example, photos taken by the mobile phone)
short_time
Silent liveness detection failed, reason: video time less than 2s
Threshold
0.4
0.5
0.6
0.7
0.8
0.9
error rate
1/10
1/100
1/1000
1/10,000
1/100,000
1/1,000,000
Name
Type
Instructions
code
int
System response code
message
string
Error Messages
request_id
string
The ID of this request
code
Field value
Description
1200
invalid argument
Invalid input parameter
2003
invalid image size
The image size (height and width in pixel) does not meet the requirements
2004
invalid content length
The image file size does not meet the requirements
2005
invalid image type or corrupted
The image type does not meet the requirements
4000
detection failed
Feature extraction failed, no face detected in the image
4004
face occlusion
The face is detected, but the eyes, nose, or mouth are partially occluded
4007
liveness silent check failed
Silent liveness detection failed
status
Description
400
BAD_REQUEST
404
NOT_FOUND
411
LENGTH_REQUIRED
413
PAYLOAD_TOO_LARGE
500
INTERNAL_ERROR