DeterecO

Our face recognition algorithm offers capabilities of convenient facial verification (one-to-one matching) and identification (one-to-many matching). The DLL is intended for biometric systems developers and can be easily integrated into the customer's security soft.

Detereco algorithm has been tested with standard face databases (BANCA, XM2VTS and others). Face model can contain multiple samples belonging to the same person. These samples should be saved with different face expressions, in different illumination conditions thus allowing to improve matching quality. For example a person could be remembered with and without beard or moustache, smiling and non-smiling etc.

15 day trial allows developers to explore the DLL’s possibilities and to try them in real applications and environments .The DLL identifies faces from almost any webcam and supports MS Windows 2003/XP/Vista/7 to run.

LicensING

The Single Computer License allows the use of the DLL on only one computer and will not be lost if computer will be reinstalled.

quantity
1
2-9
10-19
20-49
50-99
100-199
200-499
from 500

prices per single computer license
$19.00
$12.00
$7.00
$5.00
$3.00
$2.00
$1.00
$0.50

How to create good face models for fast recognition

1. Update face model with as many faces as you can:
- in different but your usual illumination conditions;
- with different but your usual face expressions;
- with different but usual distances between head and camera.
2. During process of updating of face model change distance between head and camera or even slightly turn your head right and left. During process of face recognition do the same if you are not recognized immediately.
3. Note that webcams and our recognition algorithm prefer bright illumination conditions. Avoid backlight.
4. If program recognizes you slowly or don't recognize you at all but in the past it worked fine, it is likely that there is a big change in:
- appearance (new glasses, not shaved , for women it could be a new make-up)
- illumination conditions
- distance between head and camera
I this case you should update face model with new face examples.