Complexity-based consistent-quality encoding in the cloud

Cock, Jan De; Li, Zhi; Manohara, Megha; Aaron, Anne

doi:10.1109/icip.2016.7532605

Cited by 86 publications

(104 citation statements)

References 5 publications

Supporting

Mentioning

101

Contrasting

Order By: Relevance

“…However, this fixed encoding recipe suffers from following problems: higher resolution does not always show higher quality, excessive or insufficient bitrate allocation may occur, and perceptually redundant DASH representations which are not discriminable by the human eye are often generated. Therefore, Netflix indicates that the encoding recipe should be adaptively designed Copyright © 2019 The Institute of Electronics, Information and Communication Engineers according to input video characteristics [9], [10], although their concrete methods have not been shown yet.…”

Section: Mpeg-dashmentioning

confidence: 99%

“…Similar to other research fields, introduction of the machine learning into video streaming researches is ongoing. Gaussian Mixture Model (GMM), Hidden Markov Model (HMM), Support Vector Machine (SVM), Support Vector Regression (SVR), and reinforcement learning were introduced into many components such as throughput prediction [11], rate control [12], Multi-method and image quality assessment [10], [13]. Video Multimethod Assessment Fusion (VMAF) [10] is a picture quality predictor proposed recently by Netflix that integrates multiple image quality predictors and motion information by SVR, and outputs its own score.…”

Section: Machine Learningmentioning

confidence: 99%

“…Current video streaming uses a fixed encoding recipe (bitrate-resolution pair) as shown in Table 2 for providing network adaptivity and/or device scalability by using MPEG-DASH or similar techniques. However, it has been indicated that this fixed encoding recipe suffers from several problems such as improper resolution selection and stream redundancy; therefore, the necessity of "per-title encode optimization" which adaptively generates an encoding recipe according to input video characteristics is advocated [9], [10].…”

Section: Objectivesmentioning

confidence: 99%

See 2 more Smart Citations

Methods for Adaptive Video Streaming and Picture Quality Assessment to Improve QoS/QoE Performances

Kanai

Wei

Cheng

et al. 2019

IEICE Trans. Commun.

View full text Add to dashboard Cite

Kenji KANAI †a) , Member, Bo WEI †b) , Zhengxue CHENG †c) , Student Members, Masaru TAKEUCHI †d) , Member, and Jiro KATTO †e) , Fellow SUMMARY This paper introduces recent trends in video streaming and four methods proposed by the authors for video streaming. Video traffic dominates the Internet as seen in current trends, and new visual contents such as UHD and 360-degree movies are being delivered. MPEG-DASH has become popular for adaptive video streaming, and machine learning techniques are being introduced in several parts of video streaming. Along with these research trends, the authors also tried four methods: route navigation, throughput prediction, image quality assessment, and perceptual video streaming. These methods contribute to improving QoS/QoE performance and reducing power consumption and storage size.

show abstract

Section: Mpeg-dashmentioning

confidence: 99%

Section: Machine Learningmentioning

confidence: 99%

Section: Objectivesmentioning

confidence: 99%

See 1 more Smart Citation

Methods for Adaptive Video Streaming and Picture Quality Assessment to Improve QoS/QoE Performances

Kanai

Wei

Cheng

et al. 2019

IEICE Trans. Commun.

View full text Add to dashboard Cite

show abstract

“…Note that the file size F also impacts the complexity of our proposed algorithm. However, depending on the file partitioning method and the cache contents, we may be able to operate on the chunk level instead of the bit level, where each chunk contains multiple seconds/minutes of video content [32]. Figure 4 shows an example run-time comparison between our proposed approach and the SACM approach for networks with 5 to 10 users, each with a cache of M = 3 files per user.…”

Section: B Computational Complexitymentioning

confidence: 99%

Learning to Code: Coded Caching via Deep Reinforcement Learning

Naderializadeh¹,

Asghari

2019

2019 53rd Asilomar Conference on Signals, Systems, and Computers

View full text Add to dashboard Cite

We consider a system comprising a file library and a network with a server and multiple users equipped with cache memories. The system operates in two phases: a prefetching phase, where users load their caches with parts of contents from the library, and a delivery phase, where users request files from the library and the server needs to send the uncached parts of the requested files to the users. For the case where the users' caches are arbitrarily loaded, we propose an algorithm based on deep reinforcement learning to minimize the delay of delivering requested contents to the users in the delivery phase. Simulation results demonstrate that our proposed deep reinforcement learning agent learns a coded delivery strategy for sending the requests to the users, which slightly outperforms the state-of-the-art performance in terms of delivery delay, while drastically reducing the computational complexity.

show abstract

“…Per-chunk bitrate control for consistent-quality is proposed in [5]. Instead of encoding the whole video with a target quality, the complexity of each video chunk is considered during the encoding process.…”

Section: Related Work and Backgroundmentioning

confidence: 99%

Evaluation of the Benefits of Variable Segment Durations for Adaptive Streaming

Schwarzmann

Zinner

Geissler

et al. 2018

2018 Tenth International Conference on Quality of Multimedia Experience (QoMEX)

View full text Add to dashboard Cite

Abstract-HTTP Adaptive Streaming (HAS) is the de-facto standard for video delivery over the Internet. It enables the dynamic adaptation of video quality by splitting the video clip into small segments and providing multiple quality levels per segment. Current HAS streaming services typically utilize segments of equal durations. However, this leads to video encoding overhead as segments have to start with I-frames, independently of the encoded video content. In this paper we evaluate the prospects of variable segment durations, where video segments are aligned to the video characteristics. We evaluate the reduction of the encoding overhead and investigate its impact on the stalling probability using a theoretical model. It turns out that the variable approach outperforms the fixed approach in 86% of the evaluated cases with respect to video stalls.

show abstract

Complexity-based consistent-quality encoding in the cloud

Cited by 86 publications

References 5 publications

Methods for Adaptive Video Streaming and Picture Quality Assessment to Improve QoS/QoE Performances

Methods for Adaptive Video Streaming and Picture Quality Assessment to Improve QoS/QoE Performances

Learning to Code: Coded Caching via Deep Reinforcement Learning

Evaluation of the Benefits of Variable Segment Durations for Adaptive Streaming

Contact Info

Product

Resources

About