Visionaire Documentation

Face Recognition

Nodeflux Face Recognition (Docker Stream) is a Vision Artificial Intelligence Analytics Software which provides accurate searching and recognition with searching mechanism 1:N. Using Deep Learning Algorithm to enhance the accuracy of facial inferencing, Nodeflux Face Recognition automatically performs Facial Verification and Facial Identification which is detects facial features and classifies them into recognized or unrecognized persons.
Nodeflux Face Recognition Model
Docker Stream is capable of receiving input from real-time streaming directly from an RTSP camera or recorded video. This analytic can be integrated with an external database or by creating a new database using PostgreSQL.
Face Recognition Docker Stream is suitable for surveillance cases for its capability to process streaming input. It can serve many purposes such as customer identification, blacklisted person identification, customer authentication, and many more.


Visionaire Docker Stream utilizes a combination of other services:
  1. 1.
    Postgres - For database,
  2. 2.
    Docker Snapshot - For face recognition service,
  3. 3.
    Visionaire Docker Stream - For video stream processing,
  4. 4.
    Vanilla Dashboard (Optional) - Built-in dashboard for visualization.


Model Capability

  • Due to the capability of Face Extraction based on the Deep Learning models, it is highly dependent on the characteristics of the training data and the characteristics of the real case conditions when the model is applied. There are cases where Nodeflux Face Recognition model fails to generalize similar facial features. Similar to the human eye capability, on several occasions it is failed to distinguish two people who has similar characteristics at certain angles and lighting.
  • The face biometrics model has limited capability to distinguish identical twin cases.
Identical Twin Snapshot (source: Paper)
  • The angle and lighting condition on which the camera located may affect the result of face matching.
  • The face pose must be fully capture by the CCTV camera.
  • Sunglasses and other attributed that covering the face area may reduce the accuracy of the face Biometrics.
Blurry Condition
Backlight or Low Light Condition
Object wear Sunglasses

Result Interpretation

The accuracy of the Face Recognition model highly dependents on the captured. In small cases it may present False Positive result due to several conditions such as face attributes, angle and lighting.
Blurry Object Capture
False Positive Result
Our NFV4-FR is designed to accurately identify human faces in images and videos. However, there may be rare cases where the model produces a false positive result if the image contains features that resemble a human face, such as an animal or an animated character. We strive to maintain a low error budget, but please be aware that there is a small possibility of false positive results occurring. If you have any concerns about the accuracy of the model, please don't hesitate to contact us for further assistance.