Fast transrating for high efficiency video coding based on machine learning

Van, Luong Pham; Cock, Jan De; Wallendael, Glenn Van; Leuven, Sebastiaan Van; Rodríguez‐Sánchez, Rafael; Martínez, José Luis; Lambert, Peter; Walle, Rik Van de

doi:10.1109/icip.2013.6738324

Cited by 15 publications

(8 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, a transrating technique for HEVC based on machine learning has been proposed in [29] . In this transcoder, the quantization parameters in the input HEVC video stream is changed to create a lower bit rate video.…”

Section: A Fast Transcoding Techniquesmentioning

confidence: 99%

“…In this paper, we further optimize the performance of the transcoder proposed in [29] with the following contributions.…”

Section: B Contribution Of This Papermentioning

confidence: 99%

“…Besides that, the spatial and temporal correlation is also used to support this prediction. The complexity of the transcoder can be reduced at the CU and/or PU evaluation level, contrary to only the CU level in [29] .…”

Section: B Contribution Of This Papermentioning

confidence: 99%

See 2 more Smart Citations

Efficient Bit Rate Transcoding for High Efficiency Video Coding

Van

Praeter

Wallendael

et al. 2016

IEEE Trans. Multimedia

Self Cite

View full text Add to dashboard Cite

Abstract-High efficiency video coding (HEVC) shows a significant advance in compression efficiency and is considered to be the successor of H.264/AVC. To incorporate the HEVC standard into real-life network applications and a diversity of other applications, efficient bit rate adaptation (transrating) algorithms are required. A current problem of transrating for HEVC is the high computational complexity associated with the encoder part of such a cascaded pixel domain transcoder. This paper focuses on deriving an optimal strategy for reducing the transcoding complexity with a complexity-scalable scheme. We propose different transcoding techniques which are able to reduce the transcoding complexity in both CU and PU optimization levels. At the CU level, CUs can be evaluated in top-to-bottom or bottom-to-top flows, in which the coding information of the input video stream is utilized to reduce the number of evaluations or to early terminate certain evaluations. At the PU level, the PU candidates are adaptively selected based on the probability of PU sizes and the co-located input PU partitioning. Moreover, with the use of different proposed methods, a complexity-scalable transrating scheme can be achieved. Furthermore, the transcoding complexity can be effectively controlled by the machine learning based approach. Simulations show that the proposed techniques provide a superior transcoding performance compared to the state-of-the-art related works. Additionally, the proposed methods can achieve a range of trade-offs between transrating complexity and coding performance. From the proposed schemes, the fastest approach is able to reduce the complexity by 82% while keeping the bitrate loss below 3%.Index Terms-Bit rate adaptation, complexity scalable transcoding, high efficiency video coding (HEVC), machine learning.

show abstract

Section: A Fast Transcoding Techniquesmentioning

confidence: 99%

“…In this paper, we further optimize the performance of the transcoder proposed in [29] with the following contributions.…”

Section: B Contribution Of This Papermentioning

confidence: 99%

See 1 more Smart Citation

Efficient Bit Rate Transcoding for High Efficiency Video Coding

Van

Praeter

Wallendael

et al. 2016

IEEE Trans. Multimedia

Self Cite

View full text Add to dashboard Cite

show abstract

“…On the other hand, only the highest levels in the decision tree are used in order to avoid overtraining on the training set [12]. Another method to handle the low accuracy of the machine learning model is to take the confidence of a prediction into account [10]. In that case, only predictions with a confidence above a certain threshold will be used to skip decisions in the encoder.…”

Section: Related Workmentioning

confidence: 99%

Performance analysis of machine learning for arbitrary downsizing of pre-encoded HEVC video

Van

Praeter

Wallendael

et al. 2015

IEEE Trans. Consumer Electron.

Self Cite

View full text Add to dashboard Cite

“…In order to minimize R-D efficiency losses, intelligent approaches which apply machine learning techniques to extract and analyze image characteristics and intermediate encoding results have been proposed by some authors, especially for transcoding [23,24], transrating [25] and even computational complexity reduction [26][27][28]. However, such works are still rare and do not achieve expressive complexity reductions, since they have not been fully developed and applied to multiple types of partitioning structures.…”

mentioning

confidence: 99%

Fast HEVC Encoding Decisions Using Data Mining

Corrêa

Assunção

Agostini

et al. 2015

IEEE Trans. Circuits Syst. Video Technol.

170

View full text Add to dashboard Cite

 Abstract-The High Efficiency Video Coding (HEVC) standard provides improved compression ratio in comparison to its predecessors at the cost of large increases in the encoding computational complexity. An important share of this increase is due to the new flexible partitioning structures, namely the Coding Trees, the Prediction Units and the Residual Quadtrees, with the best configurations decided through an exhaustive RateDistortion Optimization (RDO) process. In this article we propose a set of procedures for deciding whether the partition structure optimization algorithm should be terminated early or run to the end of an exhaustive search for the best configuration. The proposed schemes are based on decision trees obtained through data mining techniques. By extracting intermediate data, such as encoding variables from a training set of video sequences, three sets of decision trees are built and implemented to avoid running the RDO algorithm to its full extent. When separately implemented, these schemes achieve average computational complexity reductions of up to 50% at a negligible cost of 0.56% in terms of Bjontegaard Delta (BD)-rate increase. When the schemes are jointly implemented, an average computational complexity reduction of up to 65% is achieved, with a small BDrate increase of 1.36%. Extensive experiments and comparisons with similar works demonstrate that the proposed early termination schemes achieve the best Rate-Distortion-Complexity trade-offs among all the compared works.Index Terms-Early termination, decision trees, data mining, computational complexity, high efficiency video coding (HEVC).

show abstract

Fast transrating for high efficiency video coding based on machine learning

Cited by 15 publications

References 11 publications

Efficient Bit Rate Transcoding for High Efficiency Video Coding

Efficient Bit Rate Transcoding for High Efficiency Video Coding

Performance analysis of machine learning for arbitrary downsizing of pre-encoded HEVC video

Fast HEVC Encoding Decisions Using Data Mining

Contact Info

Product

Resources

About