Encoding and decoding architecture of checkerboard multiplexed image data
A device includes a coder or a codec configured for interleaved image data utilizing diamond shaped blocks for motion estimation and/or motion compensation and utilizing square or orthogonal transforms of residual data. In various embodiments, the decoder may be configured, among others, to perform de-blocking on edges of the diamond shaped blocks and/or data padding at boundaries of the image data. Additionally a method is proposed in which at least one of a transform and quantization process to be applied to de-multiplexed data is modified. One application is to combine left and right stereoscopic images, interleaved in a checkerboard manner.
1. A method for transforming video data, said video data further comprising source data and reference data, the method comprising:
receiving video data comprising source checkerboard data and reference checkerboard data;
de-interleaving the reference checkerboard data, said de-interleaved reference checkerboard data comprising a plurality of blocks separated according to view and parity;
buffering the de-interleaved reference checkerboard data;
de-interleaving the source checkerboard data, said de-interleaved source checkerboard data comprising a plurality of blocks separated according to view and parity;
matching the buffered de-interleaved reference data with the de-interleaved source checkerboard data to apply prediction to the de-interleaved source checkerboard data;
producing residual data from the matched data; and
transforming and quantizing said residual data.
2. The method according to claim 1 , wherein said source checkerboard data and said reference checkerboard data comprises one of a group, said group comprising: frame mode data and field mode data.
3. The method according to claim 1 , wherein the step of transforming and quantizing said de-multiplexed data further comprises modifying transforming and quantizing affecting video coding efficiency over a standardized method.
4. The method according to claim 1 , further comprising a step of at least one of motion estimation and compensation, wherein the at least one of motion estimation and compensation are performed using a block based method that does not account for overlapping blocks.
5. The method according to claim 4 , wherein the motion estimation and compensation comprise intra-prediction.
6. The method according to claim 4 , wherein motion estimation and compensation are modified apart from the transform in a manner that account for characteristics of the content.
7. The method according to claim 6 , wherein both reference and source data are re-arranged into multiple sets, each set separating data according to view and parity.
8. The method according to claim 7 , wherein the source data is matched with reference data for prediction.
9. The method according to claim 8 , wherein prediction comprises both inter and intra prediction.
10. The method according to claim 1 , wherein the data comprises field content based data and the method comprises a zig-zag scanning order.
11. A method, comprising:
receiving input image data, said input image data comprising a source data set and a reference data set, both said source data set and said reference data set further comprising a checkerboard of multiplexed data;
de-interleaving the reference checkerboard data set, said de-interleaved reference checkerboard data comprising a plurality of blocks separated according to view and parity;
buffering the de-interleaved reference checkerboard data;
de-interleaving the source checkerboard data set, said de-interleaved source checkerboard data comprising a plurality of blocks separated according to view and parity;
matching the buffered de-interleaved reference data with the de-interleaved source checkerboard data to apply prediction to the de-interleaved source checkerboard data;
producing a residual data set derived from said matched data; and
transforming and quantizing said residual data.
12. The method according to claim 11 , wherein said checkerboard of multiplexed data comprises one of a group, said group comprising: frame mode data and field mode data.
13. The method according to claim 11 , further comprising: performing transform and quantization upon said residual image data.
14. The method according to claim 11 , further comprising: performing motion estimation and compensation upon said residual image data set.
15. The method according to claim 14 , wherein the motion estimation and compensation comprise intra-prediction.
16. The method according to claim 14 , wherein motion estimation and compensation are modified apart from the transform in a manner that account for characteristics of the content.
17. The method according to claim 11 , wherein the data comprises field content based data and the method comprises a zig-zag scanning order.
18. The method according to claim 11 , wherein both reference and source data are re-arranged into multiple sets, each set separating data according to view and parity.
19. The method according to claim 18 , wherein the source data is matched with reference data for prediction.
20. The method according to claim 19 , wherein prediction comprises both inter and intra prediction.