Delay reduction for transmission and processing of video data

• US 20040042553A1
• Filed: 07/30/2003
• Published: 03/04/2004
• Est. Priority Date: 08/15/2000
• Status: Active Grant

• Current Assignee: Polycom Israel Limited (HP Inc.)
• Original Assignee: Polycom Israel Limited (HP Inc.)
• Inventors: Eshkoli, Noam, Elbaz, Moshe, Eisenberg, Aviv, Yona, Ilan
• Granted Patent: US 7,535,485 B2
• Time in Patent Office: Days
• US Class Current: 375/240.25
• CPC Class Codes: H04N 19/174 the region being a slice, e.g. a line of blocks or a group of blocks
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/40 using video transcoding, i.e. partial or full decoding of a coded input stream followed by re-encoding of the decoded output stream
• CPC Class Codes: H04N 19/61 in combination with predictive coding
• CPC Class Codes: H04N 21/234363 by altering the spatial resolution, e.g. for clients with a lower screen resolution
• CPC Class Codes: H04N 21/2662 Controlling the complexity of the video stream, e.g. by scaling the resolution or bitrate of the video stream based on the client capabilities
• CPC Class Codes: H04N 7/148 Interfacing a video terminal to a particular transmission medium, e.g. ISDN

Abstract

The present invention is a method and system for reducing delay in video communication, including, for example, video transcoding and continuous presence in a multipoint multimedia conference. The video communication control unit reduces such delay by processing a video stream in a small number of macroblocks referred to as “chunks,” without waiting to get a full frame of video data. Instead, the incoming video stream is converted into decoded chunks. These decoded chunks are transferred to an output module without waiting to decode an entire frame. An encoder in the output module encodes the decoded chunks (also referred to as encoder chunks), and transfers them to an end user without waiting for the entire frame to be processed. Thus, reducing the delay in waiting for the entire frame of video data provides improved real-time video communication.

Claims

1. A method of reducing delay of a video communication network supporting a plurality of sources, the method comprising:

receiving a compressed video data stream from an originating source;

decoding the compressed video data stream one chunk at a time to create a series of decoded chunks;

routing each decoded chunk as it is created over a common interface to an output module without waiting to decode an entire frame;

writing the decoded chunks in a memory;

retrieving encoder chunks from the memory independently from the writing;

encoding each encoder chunk; and

transmitting encoded encoder chunks as created to a target end unit without waiting to encode the entire frame.

2. A method of processing an incoming compressed data stream in an input module to improve real-time performance of a video communication system, comprising:

receiving an incoming compressed video data stream;

converting segments of the incoming compressed video data stream into decoded chunks as the segments are received;

storing pixel data corresponding to each decoded chunk in memory; and

outputting a decoded chunk while still processing other segments of a frame of the incoming compressed video data stream to which the decoded chunk belongs.

3. The method of claim 2, wherein the storing pixel data further comprises storing pixel data in a new frame memory located in the input module.

4. The method of claim 3, wherein the storing pixel data further comprises transferring the stored pixel data in the new frame memory to a reference frame memory located in the input module once all the pixel data corresponding with the frame has been stored.

5. The method of claim 4, wherein the transferring the stored pixel data in the new frame memory to a reference frame memory comprises replacing a pointer data for the new frame memory with a pointer data for the reference frame memory, and replacing the pointer data for the reference frame memory with the pointer data for the new frame memory.

6. The method of claim 4, wherein the transferring the stored pixel data in the new frame memory to a reference frame memory comprises writing the pixel data in the new frame memory over data in the reference frame memory.

7. The method of claim 2, further comprising:

scaling the stored pixel data into a scaled decoded chunk; and

outputting the scaled decoded chunk while still processing other segments of the frame of the incoming compressed video data stream to which the scaled decoded chunk belongs.

8. The method of claim 7, wherein the outputting the scaled decoded chunk further comprises transferring the scaled pixel data to a common interface.

9. A method of processing uncompressed video data in a video communication system, the uncompressed video data being available on an interface shared by a plurality of data input modules that place data onto the interface and a plurality of data output modules that extract data from the interface, the method providing an improvement in the real-time performance of the video communication system, the method comprising:

retrieving a chunk of the uncompressed video data from the interface, the chunk of the uncompressed video data comprising a portion of a video frame;

placing the chunk of the uncompressed video data into a memory element;

reading the chunk of the uncompressed video data from the memory element;

compressing the chunk of the uncompressed video data to create a compressed video data chunk; and

outputting the compressed video data chunk while a subsequent chunk of the uncompressed video data of the video frame is compressed.

10. The method of claim 9, wherein the compressing further comprises:

identifying a data rate for an intended target of the compressed video data chunk; and

compressing the chunk of the uncompressed video data in accordance with the data rate.

11. The method of claim 9, wherein the placing the chunk of the uncompressed video data into a memory comprises placing the chunk of the uncompressed video data into a specific location of a frame structure.

12. The method of claim 9, wherein the placing the chunk of the uncompressed video data into a memory and the reading are independent.

13. The method of claim 9, further comprising:

scaling the chunk of the uncompressed video data; and

compressing the scaled chunk to create a compressed scaled video data chunk.

14. The method of claim 13, further comprising outputting said compressed scaled video data chunk while the subsequent chunk of the uncompressed video data of the video frame is compressed.

15. A video processing system with improved real-time performance in processing video data streams, the video processing system comprising:

a common interface;

at least one video input module coupled to the common interface, each of the at least one video input module being communicatively coupled to a source endpoint and being operative to receive a compressed video input stream comprised of chunks from the endpoint, to decode the compressed video input stream as each chunk is received, and to transfer chunks of uncompressed data to the common interface before a full frame of uncompressed data is accumulated; and

at least one video output module coupled to the common interface, each of the at least one video output module being communicatively coupled to a target endpoint and being operative to receive the chunks of uncompressed data from the common interface, to encode the chunks of uncompressed data to create chunks of compressed video data, and to output the chunks of compressed video data for transmission to the target endpoint before a full frame of compressed video data is accumulated.