Input File Specification for Dukcapil Validation
In a sense of deep learning technology, the performance of this analytic might slightly differ in your environment, depending on several variables such as camera specs, camera height, camera angle, weather conditions, etc. We highly recommend you to test our analytics and run benchmarks on your own images, with ground truth or your quality expectations prepared beforehand. Please contact us for more info.

Input setting

Image Setting
Recommendations
Search mechanism
1:1 Matching
Minimum face area size
300 x 300 pixels, or adjust on 600 x 600 pixels
Minimum camera pixel
above .2 MP
Ideal face posture
frontal face
Photo orientation position
vertical
Accepted face tilt
>15 degree
Accepted head rotation
>30 degree
Accepted image format input
JPG or JPEG
Image file size
<800 MB (for better latency we recommend for resize the image size below 200 KB)
Total face in one frame
1 face
Image compression recommendation
bicubic, lossless

Recommended Condition

The recommended condition for each image is as follows.

Frontal Face and One Face in an Image

Conditioning

The following view configurations are acceptable but are not recommended, as they might impact analytics performance
  • Head rotation <30° (left)
head rotation <30° (left)
  • Head rotation <30° (right)
head rotation <30° (right)
In addition, there are other conditions that might affect the performance of the analytics, including (but not limited to) those listed in the following images.
  • Blurred Image
Blurred Image
  • Image with backlight
Image with Backlight
  • Wearing any type of face cover
Selfie wearing a mask
  • Wearing sunglasses
Wearing a sun glasses
  • Multiple faces within one photo
Many faces in a frame
  • Face too small and not frontal
Face image too small, cannot be identified in detail