Site News  |   Monitor Keywords  |   Monitor Archive  |   Organizer  |   Account Info  |

Video coding

Video coding description/claims

This application claims priority from application Ser. No. 10/888,702, filed Jul. 9, 2004, which is herein incorporated by reference.

The present invention relates to video coding, and more particularly to H.264 and related coding methods.

Currently, H.264 is the most advanced video compression standard and is being jointly developed by MPEG and ITU-T. It offers much higher coding efficiency compared to the existing video standards such as MPEG1, MPEG2, and MPEG4. It is widely expected that H.264 will be adopted in applications such as video conferencing, streaming video, HD-DVD, and digital video broadcasting.

In H.264 the video element bitstream is defined in the form of network abstraction layer (NAL) units. A NAL unit is a syntax structure containing an indication of the type of data to follow and bytes containing that data in the form of a raw byte stream payload (RBSP) interspersed as necessary with emulation prevention bytes. It could be a sequence header, a picture header, and a slice with header and data. In the byte stream format described in Annex B of H.264, the start code is defined as byte-aligned 0x000001 (i.e. twenty-three 0 bits followed by a single 1 bit). The byte stream format consists of a sequence of byte stream NAL unit syntax structures. Each byte stream NAL unit syntax structure contains one three-byte start code prefix (0x000001) followed by one nal_unit(NumBytesInNALunit) syntax structure. Indeed, H.264 Annex B decoding the byte stream to yield NAL units has the following steps:

1. find the next 0 byte plus three-byte start code (0x000001), and discard the 0 byte.

2. discard the three-byte start code.

3. NumBytesInNALunit is set equal to the number of bytes up to and including the last byte preceding one of: a sequence of three 0 bytes, the next start code, or the end of the byte stream.

4. NumBytesInNALunit bytes are removed from the byte stream and are decoded using the NAL unit decoding process.

5. when the next three bytes are not a start code (otherwise go to step 2) and the next four bytes are not a 0 byte plus a start code (otherwise go to step 1), repeatedly discard a 0 byte until a 0 byte plus a start code are found and then go to step 1.

The NAL unit is then decoded.

In order to avoid emulation of the three-byte start code within the NAL unit, certain rules are defined. First of all, the last byte of a NAL unit shall not be equal to 0x00. Secondly, within a NAL unit, the following three-byte sequences shall not occur at any byte-aligned position:

0x000000

0x000001

0x000002

Finally, within a NAL unit, any four-byte sequence that starts with the three bytes 0x000003 other than the following sequences shall not occur at any byte-aligned position:

0x00000300

0x00000301

0x00000302

0x00000303

• Provisional Patent
• Utility Patent

• What Is a Patent?
• What Is a Trademark or Servicemark?
• What Is a Copyright?
• Patent Laws