Enhanced Compression In Representing Non-Frame-Edge Blocks Of Image Frames

• US 20090022219A1
• Filed: 07/18/2007
• Published: 01/22/2009
• Est. Priority Date: 07/18/2007
• Status: Active Grant

• Current Assignee: NVIDIA Corporation
• Original Assignee: NVIDIA Corporation
• Inventors: Goel, Anurag
• Granted Patent: US 8,873,625 B2
• Time in Patent Office: Days
• US Class Current: 375/240.80
• CPC Class Codes: H04N 19/14 Coding unit complexity, e.g. amount of activity or edge presence estimation H04N19/146 takes precedence
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/196 being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters processing of motion vectors H04N19/513
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

Using fewer bits to indicate the prediction mode used for encoding some of the non-frame-edge blocks of a frame. In an embodiment, fewer bits are used in case of boundary blocks of a slice, or slice group. In another embodiment, fewer bits are used when adjacent blocks are encoded using inter-frame coding or switchable intra-frame coding and such adjacent block cannot be used in predicting a block.

Claims

1. A method of representing an image frame containing a plurality of source frame-edge blocks and a plurality of source non-frame-edge blocks, said method comprising:

receiving a first plurality of pixels representing said image frame in an uncompressed format, said first plurality of pixels representing said plurality of source frame-edge blocks and said plurality of source non-frame-edge blocks;

determining whether to use a common prediction mode for each of a first source non-frame-edge block and a second source non-frame-edge block based on the prediction modes used for the respective corresponding adjacent blocks, wherein said determining determines not to use said common prediction mode for each of said first source non-frame-edge block and said second source non-frame-edge block;

generating a compressed first non-frame-edge block and a second compressed non-frame-edge block using a first prediction approach and a second prediction approach respectively;

forming a compressed data by including said first compressed non-frame-edge block, said second non-frame-edge block, a first field and a second field, wherein said first field and said second field respectively identify said first prediction approach and said second prediction approach;

wherein each of said first source non-frame-edge block and said second source non-frame-edge block has equal size, and

wherein a number of bits in said first field is not equal to a number of bits in said second field.

2. The method of claim 1, wherein said generating generates a compressed non-frame-edge block by computing a difference of a corresponding source non-frame-edge block and a corresponding predicted block, wherein said corresponding predicted block is generated using a corresponding prediction approach.

3. The method of claim 2, wherein said first source non-frame-edge block is at a boundary of a slice in said image frame, wherein pixel data corresponding to outside of said slice is not used to compress the blocks contained in said slice,

wherein said first source non-frame-edge block is contained in said slice,

and wherein the number of bits in said first field is less than the number of bits in said second field.

4. The method of claim 3, wherein said slice is comprised in a slice group containing a plurality of slices.

5. The method of claim 1, wherein said forming includes a third field and a fourth field in said compressed data, wherein said third field and said fourth field respectively indicate that said common prediction mode is not used for said first compressed non-frame-edge block and said second compressed non-frame-edge block.

6. The method of claim 1, further comprising:

receiving an indication that said first source non-frame-edge block is to be encoded using an intra-prediction mode and blocks adjacent to said first source non-frame-edge-block are to be coded using an another prediction mode, wherein said forming uses different number of bits for said first field and said second field based on said indication.

7. The method of claim 6, wherein said another prediction mode comprises one of inter-prediction mode or switchable-intra prediction mode, and said indication further indicates not to use adjacent blocks for compression.

8. The method of claim 1, wherein said image frame is represented according to H.264 standard.

9. A method of decompressing an image frame, said image frame containing a plurality of frame-edge blocks and a plurality of non-frame-edge blocks, said method comprising:

receiving a sequence of bits representing said image frame in a compressed format;

examining said sequence of bits to confirm that a common prediction mode is not used for each of a first non-frame-edge block and a second non-frame-edge block, wherein said common prediction mode is one of the prediction modes used for the respective corresponding adjacent blocks, said examining also confirming that said first non-frame-edge block and said second non-frame-edge block have equal size;

determining a first set of bits indicating a prediction mode using which a first non-frame-edge block is encoded and a second set of bits indicating the prediction mode using which a second non-frame-edge block is encoded, wherein the number of bits in said first set of bits is not equal to the number of bits in said second set of bits, wherein said first set of bits and said second set of bits respectively specify a first prediction approach and a second prediction approach;

identifying the respective compressed bits representing said first non-frame-edge block and said second non-frame-edge block, wherein said first non-frame-edge block and said second non-frame-edge block have equal size;

generating a first uncompressed non-frame-edge block and a second uncompressed non-frame-edge block using said first prediction approach and said second prediction approach respectively; and

including said first uncompressed non-frame-edge block and said second uncompressed non-frame-edge block in a uncompressed data representing said image frame.

10. The method of claim 9, wherein said first non-frame-edge block is at a boundary of a slice in said image frame, wherein pixel data corresponding to outside of said slice is not used to uncompress the blocks contained in said slice,

wherein said first non-frame-edge block is contained in said slice,

and wherein the number of bits in said first field is less than the number of bits in said second field.

11. The method of claim 10, wherein said slice is comprised in a slice group containing a plurality of slices.

12. The method of claim 9, wherein said sequence of bits includes a third field and a fourth field, wherein said third field and said fourth field respectively indicate that said common prediction mode is not used for said first compressed non-frame-edge block and said second compressed non-frame-edge block.

13. The method of claim 9, wherein said sequence of bits indicates that said first source non-frame-edge block is encoded using a intra-prediction mode and blocks adjacent to said first source non-frame-edge-block are encoded using another prediction mode, wherein said determining determines the number of bits for said first field to be different from said second field based on said indication.

14. The method of claim 13, wherein said another prediction mode comprises one of inter-prediction mode or switchable-intra prediction mode, and said indication further indicates not to use adjacent blocks for decompression.

15. The method of claim 9, wherein said image frame is represented according to H.264 standard.

16. A computer readable medium storing one or more sequences of instructions for causing a system to encode an image frame, wherein execution of said one or more sequences of instructions by one or more processors contained in said system causes said system to perform the actions of:

receiving a first plurality of pixels representing said image frame in an uncompressed format, said first plurality of pixels representing said plurality of source frame-edge blocks and said plurality of source non-frame-edge blocks;

determining whether to use a common prediction mode for each of a first source non-frame-edge block and a second source non-frame-edge block based on the prediction modes used for the respective corresponding adjacent blocks, wherein said determining determines not to use said common prediction mode for each of said first source non-frame-edge block and said second source non-frame-edge block;

generating a compressed first non-frame-edge block and a second compressed non-frame-edge block using a first prediction approach and a second prediction approach respectively;

forming a compressed data by including said first compressed non-frame-edge block, said second non-frame-edge block, a first field and a second field, wherein said first field and said second field respectively identify said first prediction approach and said second prediction approach;

wherein each of said first source non-frame-edge block and said second source non-frame-edge block has equal size, and

wherein a number of bits in said first field is not equal to a number of bits in said second field.

17. The computer readable medium of claim 16, further comprising:

receiving a sequence of bits representing another image frame in said compressed format;

examining said sequence of bits to confirm that a common prediction mode is not used for each of a third non-frame-edge block and a fourth non-frame-edge block, wherein said common prediction mode is one of the prediction modes used for the respective corresponding adjacent blocks, said examining also confirming that said third non-frame-edge block and said fourth non-frame-edge block have equal size;

determining a third set of bits indicating a prediction mode using which a third non-frame-edge block is encoded and a fourth set of bits indicating the prediction mode using which a fourth non-frame-edge block is encoded, wherein the number of bits in said third set of bits is not equal to the number of bits in said fourth set of bits, wherein said third set of bits and said fourth set of bits respectively specify a third prediction approach and a fourth prediction approach;

identifying the respective compressed bits representing said third non-frame-edge block and said fourth non-frame-edge block, wherein said third non-frame-edge block and said fourth non-frame-edge block have equal size;

generating a third uncompressed non-frame-edge block and a fourth uncompressed non-frame-edge block using said third prediction approach and said fourth prediction approach respectively; and

including said third uncompressed non-frame-edge block and said fourth uncompressed non-frame-edge block in a uncompressed data representing said image frame.

18. An apparatus for processing image frames, said apparatus comprising:

means for receiving a first plurality of pixels representing a first image frame in an uncompressed format, said first plurality of pixels representing said plurality of source frame-edge blocks and said plurality of source non-frame-edge blocks;

means for determining whether to use a common prediction mode for each of a first source non-frame-edge block and a second source non-frame-edge block based on the prediction modes used for the respective corresponding adjacent blocks, wherein said determining determines not to use said common prediction mode for each of said first source non-frame-edge block and said second source non-frame-edge block;

means for generating a compressed first non-frame-edge block and a second compressed non-frame-edge block using a first prediction approach and a second prediction approach respectively; and

means for forming a compressed data by including said first compressed non-frame-edge block, said second non-frame-edge block, a first field and a second field, wherein said first field and said second field respectively identify said first prediction approach and said second prediction approach;

wherein each of said first source non-frame-edge block and said second source non-frame-edge block has equal size, and

wherein a number of bits in said first field is not equal to a number of bits in said second field.

19. The apparatus of claim 18, further comprising:

means for receiving a sequence of bits representing another image frame in said compressed format;

means for examining said sequence of bits to confirm that a common prediction mode is not used for each of a third non-frame-edge block and a fourth non-frame-edge block, wherein said common prediction mode is one of the prediction modes used for the respective corresponding adjacent blocks, said examining also confirming that said third non-frame-edge block and said fourth non-frame-edge block have equal size;

means for determining a third set of bits indicating a prediction mode using which a third non-frame-edge block is encoded and a fourth set of bits indicating the prediction mode using which a fourth non-frame-edge block is encoded, wherein the number of bits in said third set of bits is not equal to the number of bits in said fourth set of bits, wherein said third set of bits and said fourth set of bits respectively specify a third prediction approach and a fourth prediction approach;

means for identifying the respective compressed bits representing said third non-frame-edge block and said fourth non-frame-edge block, wherein said third non-frame-edge block and said fourth non-frame-edge block have equal size;

means for generating a third uncompressed non-frame-edge block and a fourth uncompressed non-frame-edge block using said third prediction approach and said fourth prediction approach respectively; and

means for including said third uncompressed non-frame-edge block and said fourth uncompressed non-frame-edge block in a uncompressed data representing said image frame.

20. The apparatus of claim 19, wherein said compressed format and said uncompressed format are according to H.264.