Home > Technology > Cloud Web-Service Architecture
Cloud Web-Service Architecture
FaceR Identity Management System: Connecting Mobile Devices with Photographic Clouds
Animetrics develops next generation identity management and biometrically enabled search engine products for authentication and search which links photographic clouds, cameras, video
and image stores via Web-service based facial search engines. Animetrics is poised for strong growth in the Government market for mobile and intelligence, and the Commercial & Consumer markets for mobile, social networking and security by revolutionizing human interfaces to computing devices and social media.
The universal deployment of mobile browsing computing devices requires trusted access and credentialing for secure communications as well the astronomical growth of cloud based photo stores requiring search and sort based on facial image content. Until now the face biometric for people on the move was not feasible enough for statistical accuracy.
Animetrics' 2D-3D FACEngine® technology provides unique technologies best suited among all face recognition technologies in industry for handling off-pose and environmentally challenging face images. The Animetrics90 SDK consists of numerous advanced client-side, server-side, service-side and storage technologies bundled into a complete package, capable of performing all tasks required for this solution, from the video and still camera interfaces, to the detection and extraction of salient faces, to the 2D-3D transformation processes, to the advanced 2D-3D matching engines, to the advanced biometric storage and retrieval systems.
Animetrics facial identity management solutions include cloud based web services:: Mobile ID for iPhone, Windows Mobile, Google Android and Blackberry users in the security marketplace. FACERTM. Animetrics is a linking partner with both Sprint and Verizon.
Web-Services for Identity and Photographic Management
The Animetrics Web-service based FaceR Identity Management System architecture has been designed to be barrier free for the consumer: any camera, any where, any time. Any mobile or fixed device which has a web browser will support the client embedded web interface which links the back end facial search engine serving the database of identities and privileges. Building a photographic cloud store requires a modern highly scalable architecture.
This diagram shows the overall architecture of the FaceR system. Thin clients communicate to the FaceR server via REST API. Application server layer is responsible for 2D image analysis, 3D avatar generation, face template generation, and any intensive computation. Application server layer is a set of clonable servers that is load balanced for throughput and scaling using Apache modproxy or Amazon ELB. The Application Server layer communicates with the MongoDB database layer to store and retrieve face templates/images/metadata. Image can also be stored on external photo stores like Picasa. Matching and searching human faces is performed via Map-Reduce and sophisticated indexing techniques. The interfaces are SOAP and REST based web services over the HTTP communication protocol, and implemented as fastcgi modules. The massively scalable nature of the consumer market place is accommodated through web server / web farming technologies for dynamically managing fastcgi servers.

The scalable nature of the database of identities to be searched is achieved via the google Map/Reduce or Microsoft Dryad architecture.