Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
173 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Bitstream Organization for Parallel Entropy Coding on Neural Network-based Video Codecs (2312.00921v1)

Published 1 Dec 2023 in eess.IV, cs.IT, and math.IT

Abstract: Video compression systems must support increasing bandwidth and data throughput at low cost and power, and can be limited by entropy coding bottlenecks. Efficiency can be greatly improved by parallelizing coding, which can be done at much larger scales with new neural-based codecs, but with some compression loss related to data organization. We analyze the bit rate overhead needed to support multiple bitstreams for concurrent decoding, and for its minimization propose a method for compressing parallel-decoding entry points, using bidirectional bitstream packing, and a new form of jointly optimizing arithmetic coding termination. It is shown that those techniques significantly lower the overhead, making it easier to reduce it to a small fraction of the average bitstream size, like, for example, less than 1% and 0.1% when the average number of bitstream bytes is respectively larger than 95 and 1,200 bytes.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (42)
  1. John W. Woods, Multidimensional Signal, Image, and Video Processing and Coding, Academic Press, Waltham, MA, second edition, 2011.
  2. Source Coding: Part I of Fundamentals of Source and Video Coding, now Publishers Inc., Hanover, MA, USA, 2011.
  3. Digital Signal Compression: Principles and Practice, Cambridge University Press, Cambridge, UK, 2011.
  4. High Efficiency Video Coding (HEVC): Algorithms and Architectures, Springer International Publishing, Switzerland, 2014.
  5. Mathias Wien, High Efficiency Video Coding: Coding Tools and Specification, Springer-Verlag, Berlin, 2015.
  6. “A new golden age for computer architecture,” Commun. ACM, vol. 62, no. 2, pp. 48–60, Feb. 2019.
  7. “The landscape of parallel computing research: A view from Berkeley,” Tech. Rep. EECS-2006-183, Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, CA, 2006.
  8. “A view of the parallel computing landscape,” Commun. ACM, vol. 52, no. 10, pp. 56–67, Oct. 2008.
  9. “Parallel entropy decoding for high resolution video coding,” in Proc. SPIE Vol. 7257: Visual Commun. Image Process., San Jose, CA, Jan. 2009.
  10. “Block Structures and Parallelism Features in HEVC,” in High Efficiency Video Coding (HEVC): Algorithms and Architectures, Vivienne Sze, Madhukar Budagavi, and Gary J. Sullivan, Eds., chapter 3, pp. 49–90. Springer, 2014.
  11. “Efficient multi-threading strategies in VVenC, an open and optimized VVC encoder implementation,” in Proc. IEEE Int. Symp. Multimedia (ISM), Naples, Italy, Dec. 2022, pp. 18–25.
  12. “Variational image compression with a scale hyperprior,” in Sixth Int. Conf. Learning Representations, Vancouver, Canada, Apr. 2018, arXiv preprint arXiv:1802.01436v2.
  13. “Scale-space flow for end-to-end optimized video compression,” in Proc. IEEE/CVF Conf. Comput. Vision Pattern Recognition (CVPR), Seattle, WA, USA, June 2020, pp. 8500–8509, IEEE/CVF.
  14. “Mobilecodec: neural inter-frame video compression on mobile devices,” in Proc. 13th ACM Multimedia Syst. Conf. (MMSys’22), Athlone, Ireland, June 2022, pp. 324–330, arXiv:2207.08338.
  15. “Hybrid spatial-temporal entropy modelling for neural video compression,” in Proc. 30th ACM Int. Conf. Multimedia, Lisboa, Portugal, Oct. 2022, pp. 1503–1511, (code at https://github.com/microsoft/DCVC).
  16. “Neural video compression with diverse contexts,” in Proc. Conf. Comput. Vision Pattern Recognition (CVPR), Vancouver, Canada, June 2023, IEEE/CVF, Code at https://github.com/microsoft/DCVC.
  17. “Compressed data organization for high throughput parallel entropy coding,” in Proc. SPIE Vol. 9599: Applicat. Digital Image Process., San Diego, CA, USA, Sept. 2015, p. 95991K, SPIE.
  18. Alistair Moffat, “Huffman coding,” ACM Comput. Surv., vol. 52, no. 4, pp. 85:1–85:35, Aug. 2019.
  19. “Parallel Huffman decoding with applications to JPEG files,” Comput. J., vol. 46, no. 5, pp. 487–497, Jan. 2003.
  20. “Accelerating JPEG decompression on GPUs,” in Proc. 28th Int. Conf. High Performance Comput. Data Analytics, Bengaluru, India, Dec. 2021, pp. 121–130, arXiv:2111.09219.
  21. “Arithmetic coding for data compression,” Commun. ACM, vol. 30, no. 6, pp. 520–540, June 1987.
  22. Amir Said, “Arithmetic coding,” in Lossless Compression Handbook, Khalid Sayood, Ed., chapter 5, pp. 101–152. Academic Press, San Diego, CA, 2003.
  23. Amir Said, “Introduction to arithmetic coding – theory and practice,” Technical Report HPL-2004-76, Hewlett Packard Laboratories, Palo Alto, CA, USA, Apr. 2004.
  24. “The use of asymmetric numeral systems as an accurate replacement for Huffman coding,” in Proc. Picture Coding Symp., Cairns, Australia, May 2015, IEEE.
  25. “Very high speed entropy coding,” in Proc. IEEE Int. Conf. Image Process., Austin, TX, USA, Nov. 1994, vol. 3, pp. 625–629.
  26. “Entropy coding in video compression using probability interval partitioning,” in Proc. Picture Coding Symp., Nagoya, Japan, Dec. 2010, pp. 66–69.
  27. “Image and video compression with neural networks: a review,” IEEE Trans. Circuits Syst. Video Technol., vol. 30, no. 6, pp. 1683–1698, June 2020, arXiv:1904.03567v2.
  28. “Auto-encoding variational Bayes,” in Int. Conf. Learning Representations, Banff, Canada, Apr. 2014, Preprint arXiv:1312.6114v10.
  29. Deep Learning, MIT Press, Cambridge, MA, 2016.
  30. “Context-based adaptive binary arithmetic coding in the H.264/AVC video compression standard,” IEEE Trans. Circuits Syst. Video Technol., vol. 13, no. 7, pp. 620–636, July 2003.
  31. Amir Said, “Efficient alphabet partitioning algorithms for low-complexity entropy coding,” in Data Compression Conf., Snowbird, UT, USA, Mar. 2005, pp. 183–192, IEEE.
  32. “Low-complexity waveform coding via alphabet and sample-set partitioning,” in Proc. SPIE Vol. 3024: Visual Commun. Image Process., San Jose, CA, USA, Feb. 1997, pp. 25–37.
  33. “High-quality, low-delay music coding in the Opus codec,” in Proc. 135th AES Conv., New York, NY, USA, Oct. 2013, Audio Engineering Society, arXiv:1602.04845.
  34. “Separating context coded and bypass coded data,” Input document JVET-O0450, Joint Video Exploration Team (JVET) of ITU-T SG16 WP3 and ISO/IEC JTC 1/SC29/WG11, Gothenburg, SE, July 2019.
  35. “Definition of the Opus Audio Codec,” Tech. Rep. RFC 6716, Internet Engineering Task Force (IETF), Sept. 2012.
  36. Peter Elias, “Universal codeword sets and representations of the integers,” IEEE Trans. Inf. Theory, vol. 21, no. 2, pp. 194–203, Mar. 1975.
  37. “Exploiting clustering in inverted file compression,” in Data Compression Conf., Snowbird, UT, USA, Mar. 1996, pp. 82–91, IEEE.
  38. G. N. N. Martin, “Range encoding: an algorithm for removing redundancy from a digitised message,” in Video and Data Recording Conference, Southampton, UK, July 1979.
  39. “An overview of the basic principles of the Q-Coder adaptive binary arithmetic coder,” IBM J. Res. Develop., vol. 32, no. 6, pp. 717–726, Nov. 1988.
  40. Michael Schindler, “A fast renormalisation for arithmetic coding,” in Data Compression Conf., Snowbird, UT, USA, Mar. 1998, p. 572, IEEE.
  41. “Arithmetic coding revisited,” ACM Trans. Inf. Syst., vol. 16, no. 3, pp. 256–294, June 1998.
  42. “MobileNVC: Real-time 1080p neural video compression on a mobiledDevice,” IEEE/CVF Winter Conf. on Applications of Computer Vision (WACV), Jan. 2024.
Citations (1)

Summary

We haven't generated a summary for this paper yet.