1. Highly Parallel Rate-Distortion Optimized Intra-Mode Decision on Multicore Graphics Processors Ngai-Man Cheung, Oscar C. Au, Senior Member, IEEE, Man-Cheung Kung, Peter H.W. Wong, Senior Member, IEEE, and Chun Hung Liu CSVT NOVEMBER 2009 1

2. Outline Introduction Intra-Prediction Parallel RD Optimized Intra-Mode Decision Experiments Conclusion 2

3. Introduction Multicore Graphics Processors ▫Graphic Processing Unit (GPUs) ▫Coprocessing units for CPUs to accelerate numerical and signal processing applications, thanks to high-performance multicore and pipeline architectures Investigate the use of GPUs to perform RD optimized intra-mode selection in AVS and H.264 3

4. Difficulties Intra-Mode Decision ▫Dependency between current block and adjacent block ▫Determine the encoding bit-rate for each of the candidate modes, some conditional branching may be needed 4

5. Contributions Analyze the dependency constraints in intra-mode decision Propose a strategy to determine the mode decisions of video blocks in parallel ▫Encode the blocks in novel orders Extend a bit-rate approximation method to estimate the rate in RD cost computation 5

6. Intra-prediction in H.264 MABCD Iabcd Jefgh Kijkl Lmnop EFGH (a) 4 × 4 current blocks and their neighboring reconstructed pixels. (b) Prediction directions and their corresponding modes. 0 5 4 6 1 8 7 3 2: DC mode 4x4 6

7. Intra-prediction in AVS 1.0 Vertical modeHorizontal mode DC modeDown-right mode Down-left mode: bidirectional prediction 8x8 7

8. Dependency Analysis Dependency constraints on block encoding order ▫Prediction Direction  Determine the RD costs of the current block is hard before all the candidate reference blocks have been encoded and reconstructed ▫Pixel Filtering(AVS)  Filtering may be applied to the reconstructed pixels of the adjacent blocks before they are used in prediction, and this filtering may involve pixels from several blocks, leading to additional block dependency 8

9. Dependency Analysis Dependency between the four 8 × 8 blocks (K1-K4) in the current macroblock and their spatially adjacent neighbor blocks (T 1-T4,L1,L2), in AVS intra-prediction 9

10. Dependency Analysis Dependency between the four 4 × 4 blocks (K1-K4)in the current 8 × 8 block and their spatially adjacent neighbor blocks(T 1-T4,L1,L2), in H.264 intra-prediction 10

11. Dependency Analysis The dependency relationships form directed acyclic graphs. ▫Parallelize the RD cost computation of the four constituent blocks of the same 16x16MB ▫Compute in parallel RD costs of the blocks from different 16x16MB 11

12. Greedy-Based Block Encoding Order Encode those blocks of which all the reference reconstructed pixels are available. 12

13. Greedy-Based Block Encoding Order AVS Example 13

14. Greedy-Based Block Encoding Order AVS Example AVS Example modify version Postpone the encoding of several blocks along the left frame boundary, All the four constituent blocks of any MB could be encoded consecutively Does not incur any execution time penalty 14

15. Greedy-Based Block Encoding Order AVS Example 15

16. Optimality Lemma 1: The proposed greedy-based encoding order can process all bottleneck path(s) P ∗ with exactly n ∗ iterations Proof of Lemma 1: ▫Suppose the greedy-based order requires more that n* iterations to process ▫At least one processing gap of length w which P* is not being processed between Ki and Ki+1 ▫There would exist an immediate parent block Bm of Ki+1 ▫Continuing with backtracking eventually one would reach some block Kj in P* ▫P1 has no processing gap >P0 ▫P* replace Po to P1 would be longer than p* P* P1 P0 16

17. Optimality Theorem 1: The proposed greedy-based order can process all the video blocks in a frame in n ∗ iterations P ∗ = {K1,K2,...,Kn ∗ }, Kn ∗, would be processed in the n ∗ th iteration by Lemma 1. Since all the paths would also end in Kn ∗, all the blocks could be processed with n ∗ iterations 17

18. Performance estimation One of the longest paths in H.264 4 × 4 intra-prediction The length can be found to be n*=((V/4)/2)x2+H/4-2 =V/4+H/4-2 n*=(V/4)x2+(H/4)/2-2 =(V/4)x2 +H/8-2 18

19. Bit-Rate Estimation Lagrangian cost function Entropy coding may involve many branching instructions, hard to implement on pipeline architecture 19

20. Fast Bit Rate Estimation for Mode Decision Tc : number of nonzero coefficients Tz : number of zeros before the last nonzero coefficients |Lk| : the absolute value of kth nonzero coefficient Fk : the frequency of kth nonzero coefficient [33] M. G. Sarwer and L.-M. Po, “Fast bit rate estimation for mode decision of H.264/AVC,” IEEE Trans. Circuits Syst. Video Technol., vol. 17, no. 10, pp. 1402–1407, Oct. 2007. 20

21. Experiments PC equipped with one GeForce 8800 GTS PCIe graphics card with 96 stream processors Intel Pentium 4 3.2 GHz processor with 1GB DDR2 memory H.264 JM 14.0 AVS RM 6.2 reference software 21

22. Encoding Bit-Rate Estimation 22

23. Parallel RD Optimized Intra-Mode Decision More than 80 times reduction QP has no significant effect H.264 23

24. Parallel RD Optimized Intra-Mode Decision Parallelism within a MB H.264 24

25. Parallel RD Optimized Intra-Mode Decision Similar speedups when RDO is disabled H.264 25

26. Parallel RD Optimized Intra-Mode Decision H.264 26

27. Parallel RD Optimized Intra-Mode Decision H.264 27

28. Parallel RD Optimized Intra-Mode Decision AVS 28

29. Parallel RD Optimized Intra-Mode Decision AVS 29

30. Parallel RD Optimized Intra-Mode Decision 39 96(processors)x2(threads)/ 5(modes) = 38.4 30

31. Conclusion Based on the dependency analysis of intra-mode decision, encode the video blocks following the greedy orders, leading to highly parallel RD cost computations. More than 80 times speedup for GPU based intra- prediction, GPU can be utilized to offload intra- prediction from CPU. To facilitate implementation on GPU, use a bitrate approximation method to estimate the rate in RD cost computation. The approximation errors only a small impact to the coding performance: no more than 0.12 dB loss in PSNR and 0.98% bit-rate increase. 31