Successfully reported this slideshow.

Community works for muli core embedded image processing

586 views

Published on

Published in: Technology, Business
  • Be the first to comment

Community works for muli core embedded image processing

  1. 1. Community works for multi-core embedded image processing Nov. 2, 2013 KELP Jeong-pyo Kong / ntekLogic jpkong@gmail.com
  2. 2. Agenda 1. Why we do this? 2. Technical backgrounds 3. Coworking with ETRI, KESSIA 4. Current our works 5. Future of our works OpenSEED imageA 2
  3. 3. 1. WHY WE DO THIS? OpenSEED imageA 3
  4. 4. IT trends are changing quite quickly OpenSEED imageA 4
  5. 5. 40 Gadgets 8 Imaging 6 Gaming α http://www.wired.com/magazine/2013/04/convergence/ Convergence OpenSEED imageA 5
  6. 6. Connected world Wearable Big data IoT/M2M http://press.teleinteractive.net/media/blogs/tialife/InternetofThingsVector.svg OpenSEED imageA 6
  7. 7. Computational world UAV Robot Autonomous Vehicle OpenSEED imageA 7
  8. 8. http://www.slideshare.net/robman/divergence-what-happens-after-convergence Divergence OpenSEED imageA 8
  9. 9. Appropriate Technology for IT business environment OpenSEED imageA 9
  10. 10. Problem solving • • • • Complex problem -> Complex solution Complex problem -> Simple solution Simple problem -> Complex solution Simple problem -> Simple solution OpenSEED imageA 10
  11. 11. http://www.youtube.com/watch?v=nrWfLAh2T3k OpenSEED imageA 11
  12. 12. What we thought • Interesting image processing and connected world • Let’s assume if we can do below things – Using improved hardware – Improving (for example) • Response time to 30% • Power saving to 15% • Accurate rate to 10% Isn’t it a big deal? OpenSEED imageA 12
  13. 13. Is it a chance for us? • Many changes are hidden from ordinary people, usually • Next generation infra • Eco system OpenSEED imageA 13
  14. 14. So, what is the result of our works? OpenSEED imageA 14
  15. 15. 1. why we do this? 2. TECHNICAL BACKGROUNDS OpenSEED imageA 15
  16. 16. Requirements for embedded image processing OpenSEED imageA 16
  17. 17. System components • Input – Image sensor – Video files on local storage – Network streaming • Processing – ISP – Low/high level image processing application • Output – Display, Signal – Network streaming OpenSEED imageA 17
  18. 18. Requirements • Low power • High performance • Easy to release : Productivity • Easy to use for user : UX/UI • Compatibility OpenSEED imageA 18
  19. 19. How can we meet these requirements OpenSEED imageA 19
  20. 20. Hardware based approach(1) • Silicon gate length : e.g. 20nm • Architecture : e.g. RISC • Instruction processing : e.g. ARM NEON OpenSEED imageA 20
  21. 21. Hardware based approach(2) • CPU trend : Operating Frequency OpenSEED imageA 21
  22. 22. Hardware based approach(3) • Multi-core Processor 1 Processor 3 Processor 2 General Purpose Processor 1 GPU Processor 4 General Purpose Processor 2 DSP Homogeneous Heterogeneous Architecture Architecture OpenSEED imageA 22
  23. 23. Software based approach • Efficient algorithm • Multi-core framework – OpenMP – OpenCL(WebCL) – CUDA • Efficient application design with multicore framework OpenSEED imageA 23
  24. 24. Heterogeneous multi-core system application OpenSEED imageA 24
  25. 25. For desktop • Games • Web acceleration • UHD video : 4K, 8K • Computer graphics, 3D, authoring tools • Financial OpenSEED imageA 25
  26. 26. For embedded system • General purpose – Games – Web acceleration – UHD video : 4K, 8K • Specialized purpose – – – – Computer vision Robot Autonomous vehicles, UAV Medical imaging OpenSEED imageA 26
  27. 27. Embedded multi-core image processing application framework OpenSEED imageA 27
  28. 28. • Future Current Desktop, smart device application HSA (Heterogeneous System Architecture) http://www.tomshardware.com/reviews/fusion-hsa-opencl-history,3262-8.html OpenSEED imageA 28
  29. 29. OpenCL(1) • Low level hardware abstraction • Supporting two parallelism methods – Data parallelization – Task parallelization • User is in charge of optimization – Engineer needs to know big picture – Regression testing… OpenSEED imageA 29
  30. 30. OpenCL(2) • Programming Model OpenSEED imageA 30
  31. 31. OpenCL(3) • Memory model OpenSEED imageA 31
  32. 32. OpenCV • Image processing library • OpenCV is being worked on supporting discrete GPU(nVidia, AMD) with x86 only, currently • OpenCV doesn’t support ARM OpenCL (Planed to release in near future) OpenSEED imageA 32
  33. 33. 2. technical backgrounds 3. COWORKING WITH ETRI, KESSIA OpenSEED imageA 33
  34. 34. ETRI providing • Platform hardware – Target board • Platform software – Kernels – Middleware – Development tools • Community base – OpenSEED OpenSEED imageA 34
  35. 35. ETRI roadmap ETRI OpenSEED imageA 35
  36. 36. KESSIA providing • Making interconnection between ETRI and community • Community operation – OpenSEED OpenSEED imageA 36
  37. 37. 3. coworking with ETRI, KESSIA 4. CURRENT OUR WORKS OpenSEED imageA 37
  38. 38. imageA = 이미자? OpenSEED imageA 38
  39. 39. imageA = image processing Acceleration OpenSEED imageA 39
  40. 40. First Trial(1) • From ETRI – Hardware : Exynos 4412 based development board. Mali-400 GPU (http://www.hardkernel.com/renewal_2011/products/prdt_info.php?g_c ode=G133888637376) – Kernel : high performance resource management kernel • Long journey for the works OpenSEED imageA 40
  41. 41. First Trial(2) • Our activities – – – – Referring user manuals in OpenSEED Kernel porting OpenCV library porting OpenCV test application development OpenSEED imageA 41
  42. 42. First Trial(3) • How about now? OpenSEED imageA 42
  43. 43. Second Trial • From ETRI – Hardware : x86 based AMD SBC – Kernel • Our activities (~2013.11) – OpenCV library – OpenCV test application development – Performance test OpenSEED imageA 43
  44. 44. 4. current our woks 5. FUTURE OF OUR WORKS OpenSEED imageA 44
  45. 45. First Trial • From ETRI – Hardware : x86 based AMD SBC – Kernel – Development tools • Our activities – OpenCV test application development – OpenCL test application – Performance test OpenSEED imageA 45
  46. 46. Second Trial • From ETRI – Hardware : ARM based development board – Kernel – Development tools • Our activities – OpenCL test application – OpenCV test application development – Performance test OpenSEED imageA 46
  47. 47. Further Trial • From ETRI – Hardware : ARM based development board – Kernel – Development tools • Our activities – – – – Existing activities + Specific application + For smart devices + Connected application : streaming and so on OpenSEED imageA 47
  48. 48. Agenda, again 1. Why we do this? 2. Technical backgrounds 3. Co-working with ETRI, KESSIA 4. Current our works 5. Future of our works OpenSEED imageA 48
  49. 49. Thank you!

×