Face Detection

The algorithm detects the presence and position of a human face (or multiple faces) on digital images and video sequences.

FD is used in all other technologies, products and applications based on the awareness of the presence and position of a face (or faces) on images or video.

View larger image

Face Features Extraction (FFE)

Face Features Extraction technology provides identification of all the key facial features (eyebrows, eyes, lips, nose, etc.) on images and video sequences.

FFE output is an input for certain other CV algorithms (including Face Features Tracking, Mimic Recognition, etc.), as well as SeeStorMe server software product for automatic avatar creation from 1 photo.

View larger image

Real-Time Head Tracking

The technology determines the spatial position of a human head as a 3D object and tracks changes of its position in real time.

Real-Time Face Features Tracking (FFT)

This technology combines Face Features Extraction and Head Tracking algorithm to track position changes of not only a head, but also the face and facial features.

FFT enables Mimic Recognition and real-time avatar animation by user’s video (3D avatar moves and behaves in sync with user). It is also used in other applications based on the awareness of the face and facial features positions in real time.

View video on YouTube

Motion Detection, Moving Object Recognition

The technologies enable various applications - from a new generation of game control interfaces to cutting-edge security and monitoring systems, e.g. Intruder Detection.

Some applications of Computer Vision technologies:

• Synthetic video:
  — automatic creation of avatars from one frontal photo to be further animated
  — avatar animation by video stream, so that avatar moves in sync with user, following his/her head and shoulder motions, and mimics
  
• Services for social networks and blogs:
  — automatic generation of photo previews containing only users' faces (while photos themselves may contain other objects, making the face hardly noticeable in preview mode), to facilitate searching people in communities
  — automatic creation of animated userpics with a human face animated by various emotional mimics, to provide users with a new visual means of self-expression in their virtual communications
  
• Image search in the Internet:
  — improving the accuracy of text content-based search results basing on visual data analysis, e.g. finding images similar to some specified sample or containing a specified object (e.g. face)
  — filtering search results basing on visual data analysis to exclude, for example, duplicates and adult content
  
• Monitoring of working conditions of various complicated systems operators (e.g. train drivers) to detect their possible weariness and distractions and prevent accidents caused by them
  
• Automatic image processing and visual improvement of human portraits (face and teeth color adjustment, visual skin healing, red eye removal, etc.)
  
• Automatic focusing on a human face - for digital photography or for making web cams and displays automatically move following user's movements to certain degrees.
  
• Video monitoring of moving objects, e.g. for intruder detection.
  
• Contactless control of computer systems by means of user's motions and gestures.