Skip to content

Main Features

The main features of the das-Face facial biometrics engine are:

  • Top-level technology: Veridas achieved top 25% worldwide in NIST FRVT 1:1
  • Verification time: around 1 second when comparing the biometric vector contained in a credential versus the face contained in an image.
  • Face liveness detection: given an image input it is possible to detect if the recorded face is authentic or it is a spoof sample. This solution is in compliance with ISO 30107-3 PAD level 1, as tested by iBeta.

To ensure optimal performance, the recommendation for images to be captured under specific conditions is needed. Veridas offers different proprietary SDKs for image photo and video recording, and they are available for multiple platforms (iOS, Android, HTML). These SDKs ensure the capture process is performed following the best conditions. The most relevant conditions are:

  • Supported image formats: JPEG/JFIF, JPEG/EXIF, JPEG Blob, PNG, and TIFF.
  • Supported video formats: MP4, WEBM, Quicktime.
  • Images must be kept as returned by the SDK, any additional compression may lead to accuracy problems.
  • Videos must be kept as returned by the SDK, any additional compression may lead to accuracy problems.
  • Face must be of at least 80px between eyes to ensure optimum verification and biometric credential computations.
  • Face must be of at least 150px width to ensure it can be processed by the face liveness detection systems. To increase accuracy, we recommended faces with more than 320px width.
  • Face is expected to be frontal with the camera in the photograph. Less than 30 degrees in head pose rotation (roll, pitch, yaw) is the optimum.
  • Face appearance must be with no significant fish-eye distortion.
  • Face movements should be smooth during video records.

This API enables the generation of biometric credentials, which are binary strings containing a biometric vector (embedding). Such vectors are a representation of the face used by das-Face for similarity comparison. Credentials have the following specifications:

  • Encoding: das-Face returns credentials encoded in base64. It is recommended to decode the string to binary before writing it to persistent storage. Any POST request containing a credential must be encoded in base64 to be understood by das-Face.
  • Length:
    • 9040 bits for models with hashes:
      • 2e0cb1f8dd755e2c9caf18c48c2b550dc7e64e57205b82c3fc5c62d6: introduced in release 3.12
      • 1f29f3d61aecf88b961d33bad141a98d9da514c8bac7c2b81eb94235: introduced in release 3.4
      • bab0b604e3cb088dd6c484a9774505f5545d3e5837874f3ec0172af5: introduced in release 3.0
    • 2896 bits for models with hashes:
      • 7c8e92b813baf87f8ec56dd840c32b4294ba14ddda436065f26ec7a0: introduced in release 3.20
      • 904fa9ef6e71ef541f20a95d3dc97821b7af43b8cd2c1bb3eb09df15: introduced in release 2.4