Page_BK_01.jpg (6305 bytes)
line_movie.gif (11715 bytes)
transparent.gif (42 bytes)

 

 

  home search contact

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

transparent.gif (42 bytes) transparent.gif (42 bytes)
transparent.gif (42 bytes)
transparent.gif (42 bytes)
transparent.gif (42 bytes)
transparent.gif (42 bytes)
transparent.gif (42 bytes)
transparent.gif (42 bytes) transparent.gif (42 bytes)

transparent.gif (42 bytes)

transparent.gif (42 bytes)

 

 

 

 

 

Image Gateway is a powerful module developed to enhance the existing Media Gateway Suite with unique indexing features based on image analysis technology. The Face Recognition and Text Title Capture modules are the first in a set, available from Pictron and their partners.  Image Gateway seamlessly provides another level of metadata information within Video Gateway.  It extracts key image features such as faces, on-screen text, colors, texture, and shapes from video key frames and uses these features as indexes for visual search.  The information appears in a metadata tab within Video Gateway making it easy to find all appearances of a specific subject just by searching on a name..  This astounding real-time face recognition analysis tool allows you to step away from the need to stop down the video and manually identify celebrities, politicians or key corporate executives.  Furthermore, important on-screen textual information, such as speaker names, stock prices, sport stats and product information is automatically indexed and matched to the relevant video providing additional text index information not found in the audio portion of the video. 

Face Recognition Features

  • Real-time human face recognition
  • Seamless integration into Video Gateway as a metadata view
  • Simple face registration process for building ‘face’ database
  • Face registration software uses a single face as the image template
  • User customizable face training and face recognition settings
  • Automatic face registration mode  dynamically builds ‘face’ database from any video
  • Automatic video segmentation based on individuals in the video

Video OCR Features
 

  • Real-time video overlay text detection and recognition
  • Seamless integration into Video Gateway as a metadata view
  • No training of text recognition required
  • Independent of text font, size, and color
  • User customizable region of interest to reduce computation load and increase accuracy
  • Recognize dark text or inverted text independent of background
  • Use customizable vocabulary to add application specific terms