1. 1 Thread-Parallel MPEG-2, MPEG4 and H.264 Video Encoders for SoC Multi- Processor Architecture Tom R. Jacobs, Vassilios A. Chouliars, and David J. Mulvaney IEEE Transactions on Consumer Electronics

2. 2 Outline Introduction Background knowledge Main purpose Previous work Methodology Experimental results Conclusions

3. 3 Introduction Background Knowledge (1/5) A number of lossy video compression standards have been developed. MPEG-1, MPEG-2, MPEG4-PART2, H.264 In order to maintain image quality and reduce bit-rates Additional computation and power consumption

4. 4 Introduction Background Knowledge (2/5) Such processing-intense consumer application algorithms are generally implemented in System-On-Chip (SOC) devices. Parallelism DLP  Data-Level Parallelism TLP  Thread-Level Parallelism

5. 5 Introduction Background Knowledge (3/5) Data-Level Parallelism (DLP) Distributing the data across different parallel processing nodes. Program: … if CPU="a" then low_limit=1; upper_limit=5 else if CPU="b" then low_limit=6; upper_limit=10 end if do i = low_limit, upper_limit Task on d(i) end do... end program

6. 6 Introduction Background Knowledge (4/5) 12 3456 7 89 10 Data array D of size 10 Processing node

7. 7 Introduction Background Knowledge (5/5) Thread-Level Parallelism (TLP) TLP is the parallelism inherent in an application that runs multiple threads at once. Benefit- Distributing the workload of a single high- performance processor among a number of slower and simpler processor cores.

8. 8 Introduction Main Purpose (1/2) Utilizing Thread-Level Parallel (TLP) techniques to improve the performance on video coding. Reduce DIC (Dynamic Instruction Count). How to improve? Workload distribution among a number of parallel-executing processors.

9. 9 Introduction Main Purpose (2/2) The results presented demonstrate that reductions in dynamic instruction count can be achieved.

10. 10 Previous Work The majority of this research is focused on coarse-granularity TLP exploitation, with distribution the workload most commonly at GOP level. GOP Multi-threading Little inter-node communication

11. 11 Previous Work In 1995, K. Shen, L. A. Rowe, and E.J. Delp implemented parallel MPEG-1 at GOP level. In 1996, S. Bozoki, S. J. P. Westen, R. L. Lagendijk and J. Biemond performed a comparison between GOP and slice level on MPEG-1.

12. 12 Previous Work In 1997, A. Bilas, J. Fritts and J. P. Singh evaluated the performance of MPEG-2 decoders using shared memory system. Akramullah, Ahmad and Liou implemented a threaded MPEG-2 encoder at the MB level by using local memory.

13. 13 Methodology Overview The threaded MPEG-2, MPEG-4 and H.264 implemented were compiled on multi-context instruction simulator (MT- ISS) based on SimpleScalar infrastructure. The most important issue Data dependancies between processors. Avoid race hazards.

14. 14 Methodology Race hazards Integer i Thread 1 0 Thread 2 1 i+1 01 12 2 Integer i Thread 1Thread 2 0 00 i+1 11 11 Race hazards Expected condition Error condition

15. 15 Methodology Thread-parallel MPEG-2 (1/5) Test model 5 (TM5) of MPEG-2 encoder is used. Computation analysis (QCIF) DIST1  52%~73% of total DIC for a search window of 6 to 62 pels respectively. FullSearch  3.5%~23.2% of total DIC. Can be improved by less complex algorithmic ME method. (such as 3-step, 4-step, diamond) FDCT, and IDCT  2.1%~21% of total DIC.

16. 16 Methodology Thread-parallel MPEG-2 (2/5)

17. 17 Methodology Thread-parallel MPEG-2 (3/5) Motion Estimation Kernel implementation can take advantage of data parallel techniques. Store the information in mbinfo structure for motion compensation. Maintain exclusivity of all variables during the parallel sections.

18. 18 Methodology Thread-parallel MPEG-2 (4/5) Forward transform FDCT first scans the MBs on a row-by-row basis, process these MBs in a row individually. Determine prediction error and applies the DCT to the block. Thread-parallel transform function can be performed in block-level.

19. 19 Methodology Thread-parallel MPEG-2 (5/5) Inverse transform IDCT scans the MBs first row-by-row and then block-by-block. Due to the absence of data dependencies between blocks  Can executed as parallel.

20. 20 Methodology Thread-parallel MPEG-4 (1/8) The implementation is based on XviD project with Advanced Simple Profile (ASP). Bidirectional frames Quarter-pel motion compensation Global motion compensation Trellis quantization Custom quantization matrices

21. 21 Methodology Thread-parallel MPEG-4 (2/8) Computation analysis (QCIF)

22. 22 Methodology Thread-parallel MPEG-4 (3/8) The nature of XivD encoder Intra-frame encoding Inter-frame encoding

23. 23 Methodology Thread-parallel MPEG-4 (4/8) Intra-frame encoding FrameCodeI (row-by-row for each MBs) Parallelize the loop for encoding the MBs in a row of the image. MB data structure  pMB. Shared memory array. The highest DIC metric in FrameCodeI is MBTransQuantIntra.

24. 24 Methodology Thread-parallel MPEG-4 (5/8) MBTransQuantIntra Forward transformation, quantization and inverse transformation. Shared data structure  pEnc Includes a count of quantization values. Serial code section. Transform specific MB pixel data into the frequency domain independently. MBPrediction and MBCoding Responsible for VLC and write to bitstream.

25. 25 Methodology Thread-parallel MPEG-4 (6/8) Inter-frame encoding FrameCodeP Part 1  Motion Estimation Part 2  Transformation  Quantization  MC

26. 26 Methodology Thread-parallel MPEG-4 (7/8) Motion Estimation Determine a MV for every MB and applies certain criteria to indicate when Intra coding should be used. Scanning in raster line order. Two kind of the process Motion prediction from current frame. ME relative to reference frames.

27. 27 Methodology Thread-parallel MPEG-4 (8/8) Motion Prediction Examining the MVs in neighbouring MBs and determining an initial estimate for ME. ● ● ● ● ● ● ● ● ● ● Ideal pattern typical pattern TLP pattern

28. 28 Methodology H.264 (1/6) Using x264 for implementation. Frame slicing Main problems of using MB-level Wide variation in processor workload. The modification of prediction algorithm is needed.

29. 29 Methodology H.264 (2/6) Slice group in H.264 A group of MBs in a frame. Can be encoded or decoded separatedly from the remainder of the frame. Not allowing motion prediction cross slice boundaries. Drawback The required bit-rate increase.

30. 30 Methodology H.264 (3/6) Comparison of different slice number

31. 31 Methodology H.264 (4/6) Comparison of different slice number

32. 32 Methodology H.264 (5/6) Different resolution with 4 slices

33. 33 Methodology H.264 (6/6) Computation analysis

34. 34 Experimental Results MPEG-2 Search Range

35. 35 Experimental Results MPEG-4 Quality Setting

36. 36 Experimental Results H.264 Quantization Parameter

37. 37 Experimental Results Comparative results

38. 38 Conclusions The DIC metric of MPEG-2, MPEG-4, and H.264 can be greatly reduced by TLP. For HD sequences, the improvement is around 84%, 92%, 96% respectively. TLP has become more significant for each new generation of video encoders.