Personal tools
You are here: Home Products F2 Face Recognition

Advanced Face Recognition from f2

Face Forensics is a highly advanced face recognition system which provides both one-to-many and one-to-one matching, as an SDK, a web service, or as a complete application.

It will:f2 Info

  • Search a database to identify an unknown face
  • Check an entire database, or multiple databases, for multiple records of the same person using different aliases
  • Detect and recognize faces in a video stream in a controlled environment
  • Verify that an individual is who they claim to be
  • Identify an individual from part of their face in a forensics or investigations environment

Face Forensics is the result of many years experience in the forefront of face recognition development. Key features are:

  • Works easily with images in existing industry- standard databases including SQL Server, Oracle, DB2, and MS Access
  • Once Face Forensics is linked to the images in an existing database it detects more than 3000 characteristics of each face. These are used to generate a unique numeric string
  • Any new face can then be encoded and its string generated and matched against the stored strings. The return is a list of images above a user-defined threshold, in order of match probability
  • Being string-based, matching is extremely fast – typically well in excess of one million faces/second. Face Forensics can be easily networked, allowing users to connect to many different databases under appropriate controls
  • Unlike many other face recognition systems, Face Forensics is available in a 64-bit version, enabling it to work with databases incorporating tens of millions of records. Text filters, e.g. on gender, can be used to narrow down the number of potential matches
  • Face Forensics accesses existing images in read-only mode so data integrity is assured. It will automatically detect and enroll new faces added to an external database without any notification from the host application.


Confirmation of an Individual’s Claimed Identity

When an individual presents an identity document such as a passport, visa, or driver’s license, f2 will confirm that the presenter’s face matches that on the document chip or server. One-to-one matching is generally an integral part of an application, so the SDK version is generally used.

Identification of Individuals

f2 can take an image of any unknown individual, encode it and match it against a database or watchlist locally or on a remote server. It returns a gallery of likely matches in order of matching probability, so that an officer can make the final decision on whether any image is of the individual or not.

Detection of Duplicates

f2 can match every facial image in a database against every other one to detect possible duplicates, thus identifying cases where an individual has multiple records under different names.

Video Screening

f2 can screen real-time or pre-recorded video. Faces in a controlled environment, i.e. forward-facing, good lighting, and high-resolution, are detected and matched against appropriate databases (offenders, missing persons, etc), with potential matches displayed for the officer or investigator to follow up. f2 can also search websites.

Forensics (partial face recognition)f2 Partial

f2 can work with part of a face and match this against the same facial region for all records in a database. By returning matches above a predefined threshold, investigators can focus their attention on a small subset of the database. This capability is unique to Face Forensics.

The Face Forensics team has over 14 years’ experience in developing face recognition technology and implementing major systems across North America and internationally.

System Requirements

Face Forensics is available as a stand-alone/networked application, as a .Net SDK, and as a web service. It runs under Windows XP, Vista, and Windows 7 (in 32-bit and 64-bit versions). It will access Oracle, SQL Server, DB2, and MS Access databases. The user interface can be switched to any language.

Face Forensics is downloadable from for evaluation for three months at no charge

Document Actions

Avon Logo  Avon - "...delivered on time, implemented smoothly..."

Centre for Addiction and Mental Health - "...not only personally rewarding and enjoyable but also have resulted in solutions..."

Citigroup - "...resulted in solutions to business requirements which are practical and cost effective..."

Trillium Logo  Credit Valley Hospital - "...could not be more pleased with Otec Solutions..."

Gordon  Group Logo  Gordon Group - "...would not hesitate to recommend Otec Solutions..."

Harlequin Logo  Harlequin Enterprises - "Our foray into a dynamic asset management system could not have been achieved in the timeline required if it had not been for Otec Solutions."

SJ  Content Logo  Pi Media, A Division of St. Joseph Corporation - "Our relationship with Otec Solutions is more like a partnership. I have peace of mind whenever we require support or consultation..."

Sears Logo   Sears Canada - "In a retail online environment, downtime is not an option. Sears is better prepared... as a result of Otec Solutions guidance."

TPS.logo.png  Toronto Police Services since 2009

Sudbury Logo  City of Sudbury since 2013

HuskyLogo.jpg  Husky Energy since 2012

CBC Logo  Canadian Broadcasting Corporation since 2003

NCC  Logo  Canada - National Capital Commission since 2011

GEH Logo  George Eastman  House since 2007

HPL Logo  Hamilton Public Library since 2005

Point Outsourcing Logo  Point Outsourcing since 2013

Sabian Logo Sabian Since 2012

RIT Logo  Rochester Institute of Technology since 2004

Tyndale U Logo  Tyndale University since 2012

U of Rochester Logo  University of Rochester since 2008

WSBC Logo WorkSafe BC  since 2013

Fuel Logo Fuel Advertising since 2001