Encoding Of An Image Frame As Independent Regions

• US 20100128797A1
• Filed: 11/24/2008
• Published: 05/27/2010
• Est. Priority Date: 11/24/2008
• Status: Abandoned Application

• Current Assignee: NVIDIA Corporation
• Original Assignee: NVIDIA Corporation
• Inventors: Dey, Soumen Kumar
• Application Number:
  US12/277,277
• Publication Number:
  US 20100128797A1
• Time in Patent Office: Days
• US Class Current: 375/240.240
• CPC Class Codes: H04N 19/103 Selection of coding mode or of prediction mode
• CPC Class Codes: H04N 19/17 the unit being an image region, e.g. an object
• 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 19/61 in combination with predictive coding

Abstract

An aspect of the present invention dynamically allocates specific blocks of an image frame to each region based on a desired number of regions and a type of regions. According to another aspect of the present invention, the values of parameters used for encoding individual regions are also dynamically computed. Due to such dynamic allocation and computations, the encoding approach may scale easily to address a large variety of image frames.

Claims

1. A machine readable medium carrying one or more sequences of instructions for causing a system to process image frames, 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 an indication that an image frame is to be encoded using a number of regions of a first type;

allocating a plurality of blocks of said image frame into said number of regions formed according to said first type;

calculating the values of a plurality of parameters after said receiving; and

encoding each of said number of regions using the calculated values for said plurality of parameters.

2. The machine readable medium of claim 1, wherein the calculated values for said plurality of parameters determine a start block in each of the regions and a sequence in which the blocks in the region are to be selected by said encoding for said first type.

3. The machine readable medium of claim 2, wherein said receiving step receives a second image frame with an indication that said second image frame is to be encoded using a second number of regions of a second type,

wherein said allocation step allocates the blocks of said second image frames into said number of regions formed according to said second type,

said calculating step calculates the corresponding values for a second plurality of parameter according to said second type,

said encoding step encodes each region of said second image frame using the calculated values for said second plurality of parameters,

wherein said second type is not the same as said first type.

4. The machine readable medium of claim 2, further comprising:

determining a number of blocks in said image frame based on a number of pixels in each dimension of said image frame and a block size used by said encoding,

wherein said allocating is performed based on said determined number of blocks and said number of regions.

5. The machine readable medium of claim 2, wherein said first type is one of raster scan, WIPE, box out,

wherein said plurality of parameters comprise a slice group change rate and a slice group change cycle,

wherein said encoding receives values for said plurality of parameters and a direction flag, said encoding determining said sequence according to said direction flag and specific blocks in each region according to said slice group change rate and said slice group change cycle.

6. The machine readable medium of claim 2, wherein said first type is interleaved type,

wherein said plurality of parameters comprise a run-length for each of said number of regions, said run-length indicating a number of a blocks in the corresponding region,

wherein said encoding determines said start block for a first region as a corner of said image frame, and the start block for subsequent regions based on the run-length of the previous regions,

wherein said encoding determines said sequence as equaling a raster scan order starting from the corresponding start block.

7. The machine readable medium of claim 2, wherein said first type is foreground/background type formed from a plurality of concentric rectangles,

wherein said plurality of parameters indicate the corners of each of said plurality of rectangles,

wherein said encoding uses a block at a corner of each of said plurality of rectangles as said start block and said sequence is determined according to a scan order.

8. A method of processing image frames, said method comprising:

receiving an indication that an image frame is to be encoded using a number of regions of a first type;

allocating a plurality of blocks of said image frame into said number of regions formed according to said first type;

calculating the values of a plurality of parameters after said receiving; and

encoding each of said number of regions using the calculated values for said plurality of parameters.

9. The method of claim 8, wherein the calculated values for said plurality of parameters determine a start block in each of the regions and a sequence in which the blocks in the region are to be selected by said encoding for said first type.

10. The method of claim 9, wherein said receiving step receives a second image frame with an indication that said second image frame is to be encoded using a second number of regions of a second type,

wherein said allocation step allocates the blocks of said second image frames into said number of regions formed according to said second type,

said calculating step calculates the corresponding values for a second plurality of parameter according to said second type,

said encoding step encodes each region of said second image frame using the calculated values for said second plurality of parameters,

wherein said second type is not the same as said first type.

11. The method of claim 9, wherein said first type is one of raster scan, WIPE, box out,

wherein said plurality of parameters comprise a slice group change rate and a slice group change cycle,

wherein said encoding receives values for said plurality of parameters and a direction flag, said encoding determining said sequence according to said direction flag and specific blocks in each region according to said slice group change rate and said slice group change cycle.

12. The method of claim 9, wherein said first type is interleaved type,

wherein said plurality of parameters comprise a run-length for each of said number of regions, said run-length indicating a number of a blocks in the corresponding region,

wherein said encoding determines said start block for a first region as a corner of said image frame, and the start block for subsequent regions based on the run-length of the previous regions,

wherein said encoding determines said sequence as equaling a raster scan order starting from the corresponding start block.

13. The method of claim 9, wherein said first type is foreground/background type formed from a plurality of concentric rectangles,

wherein said plurality of parameters indicate the corners of each of said plurality of rectangles,

wherein said encoding uses a block at a corner of each of said plurality of rectangles as said start block and said sequence is determined according to a scan order.

14. An encoder to encode a sequence of image frames, said encoder comprising:

a parameters block to receive an indication that an image frame is to be encoded using a number of regions of a first type, said parameter block to compute a corresponding number of blocks to be allocated to each of said regions based on said number, said type, a size of said image frame and a block size; and

an encoding block to receive a sequence of pixel values representing said image frame, said encoding block to encode said image frame into said number of regions, with each region containing said corresponding number of blocks.

15. The encoder of claim 14, wherein said parameters block further computes the values of a plurality of parameters after said receiving of said first type and said number of regions,

wherein said encoding block determines a corresponding set of blocks in each of the regions, a start block in each set of blocks, and a sequence of each set of blocks based on the values of said plurality of parameters and said first type.

16. The encoder of claim 15, wherein said parameter block is designed to receive an indication that a second image frame is to be encoded using a second number of regions of a second type, said parameter block computing the values for another set of parameters corresponding to said second type and also a corresponding number of blocks in each of said second number of regions,

said encoding block to encode said second image frame in said number of regions, with each region including the corresponding number of blocks, said encoding block using the values of said another set of parameters to determine a start block in each of said regions and a sequence of blocks within each region,

wherein said second type is not the same as said first type.

17. The encoding block of claim 15, wherein said parameters block computes said number of blocks in each dimension of a slice by dividing a number of pixels in the same dimension of said image frame by a number of pixels in the same dimension of each block.

18. The encoding block of claim 15, wherein said first type is one of raster scan, WIPE, box out,

wherein said plurality of parameters comprise a slice group change rate and a slice group change cycle,

wherein said encoding block is designed to receive values for said plurality of parameters and a direction flag from said parameters block, said encoding block to determine said sequence according to said direction flag and specific blocks in each region according to said slice group change rate and said slice group change cycle.

19. The encoding block of claim 15, wherein said first type is interleaved type,

wherein said plurality of parameters comprise a run-length for each of said number of regions, said run-length indicating a number of a blocks in the corresponding region,

wherein said encoding block is designed to determine said start block for a first region as a corner of said image frame, and the start block for subsequent regions based on the run-length of the previous regions,

wherein said encoding block is also designed to determine said sequence as equaling a raster scan order starting from the corresponding start block.

20. The encoding block of claim 15, wherein said first type is foreground/background type formed from a plurality of concentric rectangles as corresponding regions,

wherein said plurality of parameters indicate the corners of each of said plurality of rectangles,

wherein said encoding block is designed to use a block at a corner of each of said plurality of rectangles as said start block and said sequence is determined according to a scan order.