INTELLIGENT MULTI-STREAM VIDEO CODING FOR VIDEO SURVEILLANCE
A video surveillance system includes a plurality of cameras including an action recognition engine that classifies the video content to at least one of a plurality of predetermined actions in real time. The action recognition engine of a first camera communicates with at least one other action recognition engine, whereby detected actions and tasks related to the detected action can be exchanged. The cameras include a feature encoder operatively coupled to the action recognition engine that generates an encoded feature substream and a video encoder that receives the video content and provides an encoded video substream. A multiplexor receives the encoded feature substream and encoded video substream and generates an encoded camera bitstream including encoded video content and detected action content, which can be decoded at a receiver site for human and machine consumption.
1 . A system for video surveillance comprising a plurality of cameras, each of said cameras capturing video content, and comprising:
an action recognition engine, the action recognition engine classifying the video content to at least one of a plurality of predetermined actions, the action recognition engine having an interface enabling communication with at least one other action recognition engine, whereby detected actions and tasks related to the detected action can be exchanged with another of the plurality of cameras;
a feature encoder operatively coupled to the action recognition engine and generating an encoded feature substream;
a video encoder receiving the video content and providing an encoded video substream;
a multiplexor receiving the encoded feature substream and encoded video substream and outputting an encoded camera bitstream including encoded video content and detected action content.
2 . The system for video surveillance of claim 1 wherein upon detection of a predetermined action by a first camera, the first camera communicates at least one task to a second camera.
3 . The system for video surveillance of claim 1 wherein upon detection of a predetermined action by a first camera, the first camera communicates a first task to a second camera and a second task to a third camera.
4 . The system for video surveillance of claim 2 , wherein the tasks include at least one of object detection, object count, object tracking, and object identification.
5 . The system for video surveillance of claim 3 , wherein the objects are human and the object identification includes facial recognition.
6 . A system for video surveillance comprising a plurality of cameras, each of said cameras capturing video content, and comprising:
an action recognition engine, the action recognition engine classifying the video content to at least one of a plurality of predetermined actions, the action recognition engine having an interface enabling communication with at least one other action recognition engine, whereby detected actions and tasks related to the detected action can be exchanged with another of the plurality of cameras;
a video encoder receiving the video content and providing an encoded video bitstream;
a feature encoder operatively coupled to the action recognition engine and generating an encoded feature set therefrom;
at least one of said plurality of cameras having a feature multiplexor receiving the encoded feature sets from at least one other camera and outputting an encoded feature bitstream and detected action content for a plurality of cameras.
7 . The system for video surveillance of claim 6 wherein upon detection of a predetermined action by a first camera, the first camera communicates at least one task to a second camera.
8 . The system for video surveillance of claim 7 wherein upon detection of a predetermined action by a first camera, the first camera communicates a first task to a second camera and a second task to a third camera.
9 . The system for video surveillance of claim 7 , wherein the tasks include at least one of object detection, object count, object tracking, and object identification.
10 . The system for video surveillance of claim 9 , wherein the objects are human and the object identification includes facial recognition.