Video coding standards k. R. Rao, Do Nyeon Kim J. J. Hwang Springer 2014
Yüklə 0,84 Mb.
|
- Bu səhifədəki naviqasiya:
- Bjontegaard metric
- “Overview of AVS video standard
- AVS2 Audio and video standard China adopted as IEEE 1857.4
- THOR VIDEO CODEC
MD1. I. Irondi, Q. wang and C. Grecos, “Subjective evaluation of H.265/HEVC based dynamic adaptive video streaming over HTTP (HEVC-DASH), SPIE-IS&T Electronic Imaging, vol.9400, p. 94000B (session: real-time image and video processing), 2015.MD.P1 See [MD1]. Irondi, Wang and Grecos have presented a subjective evaluation of a HEVC-DASH system that has been implemented on a hardware test bed. In the abstract the authors state “DASH standard is becoming increasingly popular for real-time adaptive HTTP streaming of internet video in response to unstable network conditions. Integration of DASH streaming technologies with the new H.265/HEVC video coding standard is a promising area of research.” In the conclusions, they state “The results of this study can be combined with objective metrics to shed more light on the (QOE) –quality of experience- of DASH systems and correlation between the QOE results and the objective performance metrics will help designers in optimizing the performance of DASH systems”. Papers listed as references in [MD1] relate to both subjective and objective performances (each separately) of DASH systems. Combine (integrate) both objective and subjective criteria with HEVC-DASH and evaluate the QOE. Based on this evaluation optimize the performance of HEVC-DASH system. Bjontegaard metric http://www.mathworks.com/matlabcentral/fileexchange/27798-bjontegaardmetric/content/bjontegaard.m BJONTEGAARD metric calculation: Bjontegaard's metric allows to compute the average gain in PSNR or the average per cent saving in bitrate between two rate-distortions curves. See [E81], [E82], [E96] and [E198]. This metric reflects the subjective visual quality better than MSE And PSNR. Figure below explains clearly the BD-bit rate and BD-PSNR [E202].
AVS China:
[AVS8] IEEE Standards Activities Board, “AVS China - Draft Standard for Advanced Audio and Video Coding”, adopted by IEEE as standard IEEE 1857. Website: stds.ipr@ieee.org
[AVS9] D. Sahana and K.R. Rao, “A study on AVS-M standard”, Calin ENACHESCU, Florin Gheorghe FILIP, Barna Iantovics (Eds.), Advanced Computational Technologies published by the Romanian Academy Publishing House, pp. 311-322, Bucharest, Rumania, 2012. [AVS10] S. Sridhar, “Multiplexing/De-multiplexing AVS video and AAC audio while maintaining lip sync”, M.S. Thesis, EE Dept., University of Texas at Arlington, Arlington, Texas, Dec. 2010. http://www.uta.edu/faculty/krrao/dip click on courses and then click on EE5359 Scroll down and go to Thesis/Project Title and click on S. Sridhar [AVS 11] W. Gao and S. Ma, “Advanced video coding systems”, Springer, 2015.
[AVS 12] K. Fan et al, “iAVS2: A fast intra encoding platform for IEEE 1857.4”, IEEE Trans. CSVT, early access. [AVS 13] Z. He, L. Yu, X. Zheng, S. Ma and Y. He, “Framework of AVS2 Video Coding”, IEEE Int. Conf. Image Process., pp. 1515-1519. Sept. 2013. [AVS 14] S. Ma, T. Huang and C. Reader, “AVS2 making video coding smarter standards in a nutshell]”, IEEE Signal Process. Mag., vol. 32, no. 2, pp.172-183, 2015. [AVS 15] X. Zhang et al, “Adaptive loop filter for avs2”, in Proc. 48th AVS Meeting, 2014 AVS-P1 Review the references listed above on AVS China. In particular review [AVS8]. Implement various profiles in AVS China and compare its performance with HEVC using various video test sequences. Comparison needs to be based on various metrics such as PSNR, SSIM, BD-bitrate and BD-PSNR besides computational complexity. The profiles involve several projects. AVS-P2 See [AVS 12] thru [AVS 15] AVS2 is the second generation of the audio video coding standard to be issued as IEEE 1857.4 standard. It doubles the coding efficiency of AVS1 and H.264/AVC. Implement this standard and compare with HEVC based on standard performance metrics using various test sequences at different spatial/temporal resolutions. AVS-P3 See AVS-P2. In the conclusions, the authors state [AVS 12] state “Owing to their similar frameworks, the proposed systematic solution and the fast algorithms can also be applied in HEVC intra encoder design”. Explore this. AVS-P4 Fan et al [AVS12] proposed the fast intra encoding platform for AVS2 using numerous speedup methods. Using the test sequences shown in Table IV confirm the results shown in Tables V thru X and Figures 13 thru 16. AVS-P5 Compare the performance of all intra image coding techniques including JPEG series, HEVC, MPEG-4/AVC, DIRAC, DAALA, THOR, AVS1, AVS2, VP8-10 and AV1 (developed by Alliance for Open Media - AOM). As always use various test sequences at different spatial resolutions. AVS-P6 Performance comparison of SDCT with DCT and DDCT is investigated in [E392]. In AVS1 and AVS2 INTDCT is used. Replace INTDCT with INTSDCT and implement the performance comparison based on various test sequences including 4K and 8K and different block sizes. See the Tables and Figures in [E392].
Website: http://tinyurl.com/ismf77 TVC1 G. Bjontegaard et al, “ The Thor video codec”, IEEE DCC, pp. 476-485, Sun Bird, Utah, March-April 2016. Thor video codec is being developed by Cisco Systems as a royalty free codec. Details regarding block structure, intra/inter (prediction, fractional sample interpolation, motion vector coding, transform, quantization, in-loop filtering, entropy coding, transform coefficient scanning and other functionalities are described in [TVC1]). Encoder follows the familiar motion compensated, transform predictive coding as in AVC/H.264 and HEVC/H.265. For performance comparison of TVC with VP9 and x265 see Table 3 (low delay and high delay) and figures 5 and 6. For this comparison only HD sequences (classes B and E) and 1080p video conferencing sequences were used. [TVC-P1] Extend this comparison using all classes of test sequences and develop graphs based on BD-bit rate and BD-PSNR versus encoded bit rate. [TVC-P2] Consider implementation complexity as another metric and develop tables showing the complexity comparison of TVC with AVC/H.264 and HEVC/H.265 for all test sequences at various bit rates. [TVC-P3] Do extensive research and explore how the various functionalities described in [TVC1] can be modified such that the TVC can be highly competitive with AVC/H.264 and HEVC/H.265. Keep in view that the TVC is royalty free. [TVC-P4] Performance comparison of SDCT with DCT and DDCT is investigated in [E392]. In TVC, INTDCT is used. Replace INTDCT with INTSDCT and implement the performance comparison based on various test sequences including 4K and 8K and different block sizes. See the Tables and Figures in [E392]. [TVC-P4] See [E393]. Apply the FCDR to Thor video codec both as a post processing operation and as an in loop (embedded) operation. Implement the projects for TVC similar to those described in P.5.268 and P.5.269. Yüklə 0,84 Mb. Dostları ilə paylaş:
|
