News

Face Forensics Announces Disaster Victim & Mass Fatality Identification

VANCOUVER, Canada – September 9, 2024 In the event of a major disaster such as an earthquake, ship sinking, terrorist attack, etc, many bodies will be damaged and difficult or impossible to recognize. To address this f2 DVI combines capabilities developed for a range of real-world identification challenges, including:

• Full & partial face recognition
• Full & partial tattoo recognition (including scars and marks where practical)
• Corpse identification, i.e. with both eyes closed
• Identification of drowning victims

This screenshot shows a face which has been heavily damaged on the right, leaving the better side to be cropped out and positioned over a generic facial template. The missing eye position is marked relative to the good eye, face area weightings are automatically adjusted, and the search initiated. The top matches in the database are displayed on the left together with the name and Match%. Any match can be magnified for visual confirmation.

f2 DVI was originally developed for the International Committee of the Red Cross (ICRC) to help identify the bodies of migrants who had drowned in the Mediterranean trying to get to Europe, where faces may have been damaged by other boats, marine life, bloating, etc. It enables undamaged areas to be selected and eye locations (which are the critical anchor points for face recognition) to be determined. Training on capturing quality images is particularly important with shots of partial faces, where less information is available to be used. Online training sessions provided by experienced facial identification experts are available on how to optimise body/face position to maximise matching performance.

About Face Forensics Inc.
The principals of Face Forensics have been directly involved in the development of face recognition technology for over 20 years, The f2 team comprises professionals with many years’ experience in imaging technologies and applications, mathematics, biometrics, information technology, and international business.

Enquiries: Contact: info@faceforensics.com

Face Forensics Announces New Scene/Object Recognition Technology, and Major Upgrade to its Face & Tattoo Recognition Capabilities
VANCOUVER, Canada – April 19, 2022 Face Forensics has extended its image matching capabilities to finding any object in an existing database containing images, regardless of format, size, and even if only partially visible. Using an enhanced version of its crime scene matching technology together with image enhancement, objects as large as buildings and as small as tableware can be matched. Coupled with enhancements to its tattoo and face recognition algorithms, the f2 Image Recognition Suite is an indispensable addition to any agency’s forensic capabilities. The system includes automatic analysis and enrollment of the pattern and other characteristics of images in a database, automatic batch import and enrollment of images in a folder, and automatic detection and enrollment of new images that are added to external databases. Images, and the objects in them, can be of anything solid with distinguishing features, e.g. vehicles, flags, logos, furniture, etc. The software can be used stand-alone/networked or integrated into existing systems, and there is no limit to the size of the databases to be searched. Large database searches can be split across multiple processors and across a VPN. The f2 Image Recognition Suite and the modules in it are available as a stand-alone/networked application, as a .NET SDK, and as a web service. The suite is available for evaluation at no charge by government agencies and suppliers to them. Full online assistance is available.

About Face Forensics Inc
The principals of Face Forensics have been directly involved in the development of face recognition technology for over 15 years, and have already developed one of the most successful face recognition systems in North America. This has been installed in numerous police forces and achieved notable successes. The team have also been responsible for the development of one of the largest image processing systems of its type anywhere – the UK’s ChildBase system. The f2 team comprises professionals with many years’ experience in imaging technologies and applications, mathematics, biometrics, information technology, and international business.

Enquiries: Contact: info@faceforensics.com

Face Forensics Announces Major Upgrade to its Advanced Tattoo Recognition Technology
VANCOUVER, Canada – October 6, 2020 Face Forensics has significantly enhanced the tattoo encoding and matching performance of its f2 advanced tattoo algorithms, enabling images to be matched against an existing database more accurately and much faster than previously. Developed in response to demand from the law enforcement community, f2 has the ability to identify a tattoo even if it’s partially obscured, for example by clothing. The system includes automatic enrollment of existing tattoos in a database, automatic batch import and enrollment of tattoos in a folder, and automatic detection and encoding of new tattoos that are added to external databases. It can be used stand-alone or integrated into existing biometric-based systems, and there is no practical limit to the size of the databases to be searched. Large database searches can be split across multiple processors. The tattoo matching capability is available as a stand-alone/networked application, as a .NET SDK, and as a web service. f2 Tattoo Recognition is available for evaluation at no charge by government agencies and suppliers to them.

Face Forensics Inc. Announces Unique Capability to Identify Dead Bodies
VANCOUVER – March 11, 2019 Face Forensics Inc, developers of the AI-based f2 face recognition technology, has announced a significant new module – the ability to identify the faces of dead bodies, i.e. with eyes that are closed, and including where the face is significantly damaged. Closed eyes are an issue because face recognition systems place high value on the area around the eyes due to the clear edges and high contrast of the features there. If the eyeballs are not visible as anchor points the recognition performance can be severely degraded or simply not work at all. Called Corpse-ID, the new module builds on Face Forensics’ unique partial face recognition capability to enable an investigator to estimate the eye center locations optimally. This capability also allows an image of an undamaged part of a face, for example one side or just the eyes, to be matched against the same area on previously enrolled full faces. Corpse-ID adds another unique capability to Face Forensics’ f2 Image Recognition Suite, which also includes tattoo recognition (also with partials) and scene recognition.

Face Forensics Inc. Announces 300% Face/Tattoo Enroll Speed Increase
VANCOUVER – May 21, 2017 Vancouver-based Face Forensics Inc, developers of the highly-advanced f2 face and tattoo recognition technology, has announced a substantial increase in its image enrollment speed. While f2 always worked in 64-bit on the server side, this new release extends 64-bit to the client side as well. This together with other major enhancements has more than tripled the image enrollment speed. f2’s use of modular hardware architecture plus its full use of multithreading and multiple cores and processors, results in search speeds of faces and tattoos that can be in the many millions of images per second, and delivers scalability from one-to-one matching for access control through to one-to-many matches for the largest databases.

About Face Forensics
The principals of Face Forensics have been directly involved in the development of face recognition technology for over 15 years, and have already developed one of the most successful face recognition systems in North America. This has been installed in numerous police forces and achieved notable successes. The team have also been responsible for the development of one of the largest image processing systems of its type anywhere – the UK’s ChildBase system. The f2 team comprises professionals with many years’ experience in imaging technologies and applications, mathematics, biometrics, information technology, and international business.

Enquiries: Contact: info@faceforensics.com

Face Forensics Announces New Advanced Identity Verification Capability
VANCOUVER, Canada – October 4, 2016 Face Forensics’ advanced face recognition technology can be used in both identification and identity verification applications. For its primary purpose is to verify that the person presenting an ID card is the person to whom it was issued, and thus prevent unauthorized individuals from entering the building, and one employee entering on behalf of another. When an individual presents their personal ID card at an entry point, f2 will immediately determine if the presenter’s face matches the face of the owner of the card. As the individual walks through the entrance of a building towards the camera, the employee’s ID card number is read, either through a card reader or by reading a chip on the card, validated, and used to retrieve the individual’s photo from the employee database. In parallel a video camera will automatically take several shots of the individual’s face. Each shot will be encoded, with each encode array compared with ones for this individual in the database. If the match% for any face exceeds a user-defined acceptance threshold the two faces are deemed to match and the person is accepted to be who they claim to be. At this point the employee will be permitted to enter the building. If there’s no match they won’t be allowed to enter and security will take them to one side for further checking. If they refuse to cooperate f2 can be used in identification mode to compare the camera image with the faces of all the employees in the database in a one-to-many match to determine if the individual is an employee, ex-employee, or an unknown person. The system will also detect buddy-punching, where one employee tries to enter on behalf of another, and identify both of them. In all cases the application can keep a time-stamped record of the images as well as details of the specific location that the match took place, for future investigation if required. Where a face is intentionally partially obscured, e.g. by the person’s hand, f2’s unique partial face recognition capability can be used to help identify the individual from the features that are visible.

Example of 1:1 Match

Law & Order Magazine Features Face Forensics Tattoo Recognition as Cover Article
VANCOUVER, Canada – July 28, 2016 The cover article in the June 2016 edition of Law & Order magazine was an article on the company’s new and unique tattoo recognition capability written by Face Forensics CEO Iain Drummond. Requested by the magazine, the article focused particularly on its use in a police environment. The article can be seen at: http://www.hendonpub.com/resources/article_archive/results/details?id=5807 Tattoo recognition is a specific application of Face Forensics’ image recognition technology, which was initially applied to face recognition and our partial face recognition capability, which is also unique. The software can handle heavily-cropped tattoos and includes optional color and text filters to reduce the number of possible matches when dealing with large databases. Tattoo recognition is a particularly challenging task as unlike faces a tattoo can be of virtually anything, so there are no obvious anchor points around which to determine the relative position of other features. Solving this and other associated challenges has resulted in f2 being the only generally-available tattoo recognition system in the world..

Face Forensics Announces World’s First Generally-Available Tattoo Recognition System
VANCOUVER, Canada – October 24, 2015 Face Forensics has significantly expanded its f2 face recognition technology by adding the capability to match images of tattoos against databases containing them. It’s been developed in response to demand from the law enforcement community, who have asked for the ability to identify a tattoo or other image, even although it’s been cropped, reformatted, or is just similar. Although this capability is a new module within the Face Forensics Recognition Suite it is fully integrated so that it can connect directly back to the face and name of the owner of any scar, mark, or tattoo. The system includes automatic enrollment of existing tattoo images, and automatic detection and encoding of new tattoos that are added to external databases. Face Forensics accesses these images in read-only mode so data integrity is assured. The Tattoo module is an independent but integrated module within the f2 image recognition suite, which includes both full face recognition and partial face recognition – another unique and powerful capability. It will work alongside existing biometric systems. The tattoo module is available within the f2 Image Recognition Suite as a stand-alone/networked application, as a .Net SDK, and as a web service. The f2 Image Recognition Suite is available for evaluation by interested organizations and suppliers to them.

Face Forensics Inc. Announces High Speed Partial Face Recognition Capability
VANCOUVER – June 22, 2015 Face Forensics Inc, developers of the highly-advanced f2 face recognition technology, has announced a significant enhancement to its unique partial face recognition capability. This capability allows an image of part of a face, for example one side or just the eyes, to be matched against the same area on previously enrolled full faces. Databases are searched at extremely high speed – f2 can be configured to match well in excess of 1m faces/second. The top matches are displayed as thumbnail images in order of match %, making it straightforward to determine if the full face is already in the database. Face Forensics’ unique partial face recognition capability is invaluable for recognizing faces which are partially-concealed, burnt or injured, as well as criminals wearing balaclavas, etc.

Face Forensics Partial Face Recognition Technology Featured on Sky News
VANCOUVER – October 8, 2014 In late September London-based Sky News ran a prime-time feature on the identification of individuals who hide their face by wearing balaclavas. Central to this feature was the use of Face Forensics’ unique partial face recognition technology. Unlike other face recognition systems, all of which require a full facial image in order to match a face, f2 can work with just a section of a face.
This is invaluable where the face is partially concealed by, a building, another person, or by clothing, so that just one eye can be seen. f2 can also work with a “letterbox” view where only the area around the eyes is visible. The partial face recognition capability adds a new and invaluable tool to security and law enforcement agencies around the world.

Face Forensics Inc. Announces Partnership with Otec Solutions Inc.
VANCOUVER – May 16, 2014 Face Forensics Inc, developers of the highly-advanced f2 face recognition technology, has announced a business partnership with Toronto-based Otec Solutions Inc.

Under the agreement, Otec will incorporate Face Forensics’ f2 technology into its Digital Evidence Management Solution (DEMS). When set up, f2 automatically enrolls all the faces in selected databases by determining over 3000 features in the central area of the face and transforming the resulting data into a unique numeric string for each face. This enables facial searches to be conducted across all databases at extremely high speed – well in excess of 1m faces/second. The top matches are displayed as thumbnail images in order of match %, making it straightforward for an officer to visually scan the top matches to determine if the face is already known. Integrating this capability with Otec’s Digital Evidence Management enables law enforcement agencies to securely access and search data, including faces, across a wide range of applications and devices.

Otec’s Digital Evidence Management Solution streamlines ingestion of crime scene photography and video with the Camera Wizard. DEMS also provides a robust, fast search interface for ordering evidence for trial. Orders are processed through a forensics workflow to deliver orders to police, defense and prosecutors. Billing for these orders is viable. About Otec Solutions Inc.

Otec Solutions Inc. is a privately held systems integrator with a focus on providing Digital Evidence Management, Digital Asset Management, Automated Workflow & secured remote access of those services based on the individual requirements of its clients. Otec markets, supports and customizes the world’s leading Digital Asset Management software, Cumulus by Canto. As a Canto Certified Systems Integrator and Platinum Level Sales Partner, Otec is well-equipped to adapt it to meet specific needs. Otec holds certifications in critical IT disciplines to offer a complete and comprehensive design, installation, training and support of the entire system infrastructure where required.

Contact: info@otecsolutions.com Website: http://www.otecsolutions.com

Face Forensics Announces Significant Enhancements to its Eye-Detection Capabilities
Vancouver - April 29, 2014 The performance of Face Forensics’ f2 facial recognition technology has been significantly enhanced by the upgrading of its eye detection logic. Eye detection is critical to the performance of face recognition systems as it performs a vital initial role by finding the key anchor points around which, in f2’s case, the more than 3000 characteristics of the face are determined. This improvement means that eye centers can be more accurately located even when the facial image is not particularly clear. This development complements Face Forensics’ partial face recognition capabilities, enabling images of both full and partial faces to be readily identified. F2 search speeds are extremely fast (well in excess of 1m faces/second). It is available in both 32 and 64 bit, so is well able to handle extremely large databases.

Forensic Protection Inc. & Face Forensics Agree to Partner on Image Enhancement and Face Recognition
VANCOUVER - December 19th 2012 – With increasing demand from government agencies around the world for the ability to recognize wanted individuals from their face, often from CCTV video streams, there’s a need to be able to enhance the quality of what are very often low-resolution and poor images. Forensic Protection Inc. has developed technology which, unlike most providers of this capability, does this automatically. This is provided as a service – customers send in the video (or audio) clips they want enhanced, and Forensic Protection processes and returns them. Where these are of faces, Face Forensics’ advanced face recognition capabilities can be used to match the enhanced image against a known image of the suspect, or a database of known offenders, to help identify them. This includes the unique ability to match part of a face, i.e. one which is injured, or partially hidden by another person, a building, or even a balaclava. From Face Forensics’ point of view the partnership offers its customers the ability to recognize faces where otherwise this would not be feasible, opening a range of possibilities which previously did not exist. The Face Forensics f2 system is based on many years’ experience in developing face recognition technology for law enforcement, both in the US and overseas. It offers both one-to-one and one-to-many matching, is unaffected by age, race, or color, and operates at a search speed of over 1m faces/second.

Police Central & Face Forensics Announce Partnership to add Advanced Face Recognition Capabilities to Police Central’s Jail Management System.
VANCOUVER – May 1, 2012 – Police Central’s comprehensive Jail Management System is being enhanced with the addition of advanced face recognition technology. This will facilitate the recognition of unidentified offenders, including those who refuse to give their name when arrested. It will also identify individuals with multiple records under different aliases. Atlanta-based PCI’s customer base includes approximately eighty-five law enforcement agencies in ten states, including the City of Atlanta Police Department and the City of Atlanta Department of Corrections. PCI’s web-based Law Enforcement Software suite already comprises thirty-five modules covering all aspects of law enforcement. The face recognition capability will complement and add value to these. The Face Forensics system is based on many years’ experience in developing this type of technology for law enforcement, both in the US and overseas. It is specifically designed to be easy to integrate with systems like PCI’s JMS. Recognition performance is unaffected by age, race, or color, and offers a search speed of over 1m faces/second. A police department can search the databases of adjacent police departments under appropriate controls. Face Forensics is the only system in the world to provide partial face recognition – offering the ability to recognize injured faces, criminals wearing balaclavas, etc. The Face Forensics team have also been responsible for the development of one of the largest pedophilic image recognition systems in the world – the UK Serious Organized Crime Agency’s ChildBase system.

Face Forensics Announces Full Support for SDK and Web Services Versions of f2
VANCOUVER - January 18, 2012 - Face Forensics’ facial recognition technology is now available as a standalone/networked application, an SDK, and as a web service. Face Forensics is unique in being the only facial recognition system to include partial face recognition, and is specifically designed to connect to existing databases. The f2 search speeds are extremely fast (well in excess of 1m faces/second). It is available in both 32 and 64 bit, so is well able to handle extremely large databases.

Vancouver Riots and Face Forensics.
VANCOUVER – June 21, 2011 – Face Forensics Inc, the developer of the advanced f2 face recognition technology, has provided Global News with insights on how f2 could be used to analyze imagery and video taken from the Vancouver riots. See the interview here: Global News interview

Face Forensics Inc. Announces automatic matching of faces across multiple databases
VANCOUVER – March 24, 2011 – Face Forensics Inc, developers of the highly-advanced f2 face recognition and unique partial face recognition technology, has announced the ability to automatically match all faces in multiple existing databases to identify individuals operating under different names.
f2 can automatically locate and enroll facial images in SQL Server. It will take each enrolled face and match it against every other face in the same database or in different databases. Thus different records of the same individual operating under different names will be identified. All matches above a user-defined threshold are written to a separate file for an officer to confirm. By eliminating all but the most likely matches, f2’s database matching capabilities transform what is in practice an impossible task into something that is straightforward and fast for an officer to undertake.
f2 is specifically designed to work easily and quickly with existing databases. It will detect single or multiple faces in a photo, video, or website, and match them in real time. It is available as a stand-alone or networked application and as an SDK, and has the capability to work with extremely large databases. An evaluation copy can be downloaded from www.faceforensics.com

Face Forensics Inc. Announces World First: Partial Face Recognition
VANCOUVER – November 11, 2010 – Face Forensics Inc, developers of the highly-advanced f2 face recognition technology, has announced the unique ability to match part of a face against a database of faces. This is particularly useful in a forensics environment where only body parts are left, e.g. after an explosion or fire, or with surveillance imagery where only part of a face is available. The image of the facial part is positioned on a generic facial template. Controls are available to resize the partial face as well as adjust the key characteristics of the generic template to the partial face. Once positioned optimally, the partial face is encoded. This encoding is then matched against the same areas of full faces in the database. Matches above a predefined threshold are displayed for an investigator to review. f2 is specifically designed to work easily with existing databases. It will detect single or multiple faces in a photo, video, or website. It is available both as a free-standing application and as an SDK, and has the capability to work with extremely large databases. An evaluation copy can be downloaded from www.faceforensics.com

360 Biometrics to Market Face Forensics' Face Recognition Technology
VANCOUVER – July 21, 2010 – Face Forensics Inc, the developer of the advanced f2 face recognition technology, has signed an agreement with Silicon Valley based biometric solution reseller 360 Biometrics to market Face Forensics in its markets around the world. 360 Biometrics plans to promote both the one-to-one and one-to-many versions of Face Forensics. These systems are available as both stand-alone applications and SDKs. 360 Biometrics is also interested in Face Forensics’ newly announced capability to identify partial faces to be integrated into 360 Biometrics' security solutions. Designed principally for use by forensics laboratories, partial face recognition is a unique capability only available from Face Forensics.

About 360 Biometrics Inc.
360 Biometrics is based in Sunnyvale, California, and provides a complete range of biometric solutions including software, hardware and consulting. They are also an established integrator and value-added reseller of multiple biometric technologies and products. The focus of 360 Biometrics is to assist small businesses, large enterprises and government bodies with the identity and security solutions using various biometric technologies such as fingerprint, face recognition, iris, etc. The customer base of the company ranges from small businesses to very large commercial enterprises and Government organizations at both the State and Federal level. 360 Biometrics also serves all markets worldwide.

About Face Forensics Inc. The principals of the company have already developed one of the most successful face recognition systems in North America. This has been installed in numerous police forces and other locations and achieved major successes. The team have also been responsible for the development of the largest pedophilic image processing system anywhere – the UK’s ChildBase system.

The f2 team comprises professionals with many years combined experience in imaging technologies and applications, mathematics, biometrics, information technology, and international business.
Enquiries: Contact: info@faceforensics.com