Method and apparatus for efficiently prioritizing elements in a video stream for low-bandwidth transmission

1. A method for prioritizing packets from a plurality of different content sources in a communications network, the method comprising:

at each content source from the plurality of different content sources, mapping content to a content identifier according to a content type of the content, the content including a representation of a real-world article, the content type indicating substantive subject matter contained within audio, image or video data of the content;

at each content source, assigning the content identifier as a function of the content type of the content in a packet of a packet stream on a packet-by-packet basis; and

at each content source, forwarding the content identifier in a codec-neutral format in a Real-Time Protocol (RTP) header extension with the packet across the network to enable a downstream network node which is independent of the content sources receiving the packet to effect prioritization of the packet received within the packet stream, irrespective of codec type, by extracting the content identifier forwarded in the codec-neutral format from the packet received to determine the priority for the packet according to a content mapping table including mappings between content types, content identifiers and priorities,

wherein each content identifier includes a respective field specifying a respective version of the content identifier;

at the downstream network node, extracting content identifiers from the packets generated from the different content sources, wherein a priority is assigned to each extracted content identifier according to the content mapping table, the assigning of a priority to each extracted content identifier including decoding the respective version included in each extracted content identifier;

at the downstream network node, determining drop or forward actions for the packets prioritized based on the assigned priority values.

2. The method according to claim 1 , wherein each content source is a video camera, a video recorder, or a gateway.

3. The method according to claim 1 , wherein the content is contained within a video frame.

4. The method according to claim 1 , wherein the substantive object identifier indicates a face, a license plate, or a personalized identifier.

5. The method according to claim 1 , wherein the content mapping table is a standard content mapping table.

6. The method according to claim 5 , wherein the downstream node assigns prioritization using the standard content mapping table.

7. The method according to claim 1 , further including assigning a priority to the content identifier.

8. The method according to claim 7 , wherein mapping content to the content identifier occurs at the content source and assigning the priority to the content identifier occurs at a device separate from the content source.

9. The method according to claim 1 , wherein forwarding the content identifier with the packet includes forwarding the content identifier in a packet associated with the packet including the content.

10. The method according to claim 1 , further comprising mapping video content to a substantive object through the use of the content identifier specified in the Real-Time Protocol (RTP) header extension maps.

11. The method according to claim 1 , wherein packets within the packet stream are selectively forwarded or dropped on a network according to the prioritization of the packets, which is based on the content identifier of each packet, to forward useful security-pertinent content or drop content less pertinent to security respectively.

12. The method according to claim 1 , wherein the substantive object identifier identifies the substantive subject matter contained within content comprising a scene in an image.

13. The method according to claim 1 , wherein the substantive subject matter comprises an object in an image or video data of the content captured from a content source comprising a camera, a priority of an object having a distance closer to the camera being higher than an object having a distance farther away from the camera.

14. A method for prioritizing packets from a plurality of different content sources at a network node which is independent of the content sources in a communications network, the method comprising:

receiving packets at the network node from different content sources over a network, the received packets having a set of content identifiers forwarded in a codec-neutral format in a Real-Time Protocol (RTP) header extension and being a function of respective content type of content included in the received packets, the content including a representation of a real-world article, wherein each content identifier includes a respective field specifying a respective version of the content identifier;

accessing a content mapping table including mappings between a plurality of content types, content identifiers and prioritization values, each content type indicating substantive subject matter contained within audio, image or video data of content;

extracting content identifiers from the packets generated from the different content sources, wherein a priority is assigned to each extracted content identifier according to the content mapping table, the assigning of a priority to each extracted content identifier including decoding the respective version included in each extracted content identifier; and

determining drop or forward actions for the packets prioritized based on the assigned priority values.

15. The method according to claim 14 , wherein the content is contained within a video frame.

16. The method according to claim 14 , further including assigning the set of content identifiers at a video source.

17. A system comprising:

a plurality of different content sources configured to communicate over a communications network, each content source including at least a processor and a memory storing computer-executable instructions which, when loaded and executed by the processor, configured to implement:

a video content classifier configured to assign a content identifier as a function of a content type of content in a packet of a video packet stream on a packet-by-packet basis, the content including a representation of a real-world article, the content type indicating substantive subject matter contained within video data of the content in a packet of a video packet stream; and

a forwarding engine configured to forward the content identifier with the packet in a codec neutral format in a Real-Time Protocol (RTP) header extension across a network to enable a downstream network node which is independent of the content sources to effect prioritization of the packet within the video packet stream, irrespective of codec type, by extracting the content identifier forwarded in the codec-neutral format from the packet received to determine the priority for the packet according to a content mapping table including mappings between content types, content identifiers and priorities,

wherein each content identifier includes a respective field specifying a respective version of the content identifier,

at the downstream network node, content identifiers are extracted from the packets generated from the different content sources, wherein a priority is assigned to each extracted content identifier according to the content mapping table, the assigning of a priority to each extracted content identifier including decoding the respective version included in each extracted content identifier, and

at the downstream network node, drop or forward actions are determined for the packets prioritized based on the assigned priority values.

18. A system comprising:

a plurality of different content sources configured to communicate over a communications network, each content source including at least a processor and a memory configured to:

process a video scene and produce a packet stream including a plurality of packets representing the video scene;

assign a content identifier to each packet in the plurality of packets, each packet including an assigned content identifier, the content identifier assigned as a function of a content type of the video content represented by each packet, the content including a representation of a real-world article, the content type indicating substantive subject matter contained within video data of video content in a packet; and

forward each assigned content identifier with each corresponding packet in a codec neutral format in a Real-Time Protocol (RTP) header extension across a network to enable a downstream network node which is independent of the content sources to effect prioritization of the corresponding packet within the packet stream, irrespective of codec type, by extracting the assigned content identifier forwarded in the codec-neutral format from the packet received to determine a priority for the corresponding packet according to a content mapping table including mappings between content types, content identifiers and priorities,

wherein each content identifier includes a respective field specifying a respective version of the content identifier,

at the downstream network node, content identifiers are extracted from the packets generated from the different content sources, wherein a priority is assigned to each extracted content identifier according to the content mapping table, the assigning of a priority to each extracted content identifier including decoding the respective version included in each extracted content identifier, and

at the downstream network node, drop or forward actions are determined for the packets prioritized based on the assigned priority values.