Minh Nguyen (AAU, Austria), Ekrem Çetinkaya (AAU, Austria), Hermann Hellwagner (AAU, Austria), and Christian Timmerer (AAU, Austria)
Abstract: Recently, mobile devices have become paramount in online video streaming. Adaptive bitrate (ABR) algorithms of players responsible for selecting the quality of the videos face critical challenges in providing a high Quality of Experience (QoE) for end users. One open issue is how to ensure the optimal experience for heterogeneous devices in the context of extreme variation of mobile broadband networks. Additionally, end users may have different priorities on video quality and data usage (i.e., the amount of data downloaded to the devices through the mobile networks). A generic mechanism for players that enables specification of various policies to meet end users’ needs is still missing. In this paper, we propose a weighted sum model, namely WISH, that yields high QoE of the video and allows end users to express their preferences among different parameters (i.e., data usage, stall events, and video quality) of video streaming. WISH has been implemented into ExoPlayer, a popular player used in many mobile applications. The experimental results show that WISH improves the QoE by up to 17.6% while saving 36.4% of data usage compared to state-of-the-art ABR algorithms and provides dynamic adaptation to end users’ requirements.
Keywords: ABR Algorithms, HTTP Adaptive Streaming, ITU-T P.1203, WISH
Vignesh V Menon (Alpen-Adria-Universitat Klagenfurt); Hadi Amirpour (Alpen-Adria-Universität Klagenfurt); Christian Timmerer (Alpen-Adria-Universität Klagenfurt, Austria); Mohammad Ghanbari (University of Essex, UK).
Abstract:
High Efficiency Video Coding (HEVC) improves the encoding efficiency by utilizing sophisticated tools such as flexible Coding Tree Unit (CTU) partitioning. The Coding Unit (CU) can be split recursively into four equally sized CUs ranging from 64×64 to 8×8 pixels. At each depth level (or CU size), intra prediction via exhaustive mode search was exploited in HEVC to improve the encoding efficiency and result in a very high encoding time complexity. This paper proposes an Intra CU Depth Prediction (INCEPT) algorithm, which limits Rate-Distortion Optimization (RDO) for each CTU in HEVC by utilizing the spatial correlation with the neighboring CTUs, which is computed using a DCT energy-based feature. Thus, INCEPT reduces the number of candidate CU sizes required to be considered for each CTU in HEVC intra coding. Experimental results show that the INCEPT algorithm achieves a better trade-off between the encoding efficiency and encoding time saving (i.e., BDR/∆T) than the benchmark algorithms. While BDR/∆T is 12.35% and 9.03% for the benchmark algorithms, it is 5.49% for the proposed algorithm. As a result, INCEPT achieves a 23.34% reduction in encoding time on average while incurring only a 1.67% increase in bit rate than the original coding in the x265 HEVC open-source encoder.
Keywords:HEVC, Intra coding, CTU, CU, depth decision
ViSNext’21: 1st ACM CoNEXT Workshop on Design, Deployment, and Evaluation of Network-assisted Video Streaming
In recent years, we have witnessed phenomenal growth in live video traffic over the Internet, accelerated by the rise of novel video streaming technologies, advancements in networking paradigms, and our ability to generate, process, and display videos on heterogeneous devices. Regarding the existing constraints and limitations in different components on the video delivery path from the origin server to clients, the network plays an essential role in boosting the perceived Quality of Experience (QoE) by clients. The ViSNext workshop aims to bring together researchers and developers working on all aspects of video streaming, in particular network-assisted concepts backed up by experimental evidence. We warmly invite submission of original, previously unpublished papers addressing key issues in this area, but not limited to:
Design, analysis, and evaluation of network-assisted multimedia system architectures
Optimization of edge, fog, and mobile edge computing for video streaming applications
Optimization of caching policies/systems for video streaming applications
Network-assisted resource allocation for video streaming
Experience and lessons learned by deploying large-scale network-assisted video streaming
Internet measurement and modeling for enhancing QoE in video streaming applications
Design, analysis, and evaluation of network-assisted Adaptive Bitrate (ABR) streaming
Network aspects in video streaming: cloud computing, virtualization techniques, network control, and management, including SDN, NFV, and network programmability
Routing and traffic engineering in end-to-end video streaming
Topics at the intersection of energy-efficient computing and networking for video streaming
Network-assisted techniques for low-latency video streaming
Machine learning for improving QoE in video streaming applications
Machine learning for traffic engineering and congestion control for video streaming
Solutions for improving streaming QoE for high-speed user mobility
Analysis, modeling, and experimentation of DASH
Big data analytics at the network edge to assess viewer experience of adaptive video
Reproducible research in adaptive video streaming: datasets, evaluation methods, benchmarking, standardization efforts, open-source tools
Novel use cases and applications in the area of adaptive video streaming
Advanced network-based techniques for point clouds, light field, and immersive video
Jesús Aguilar Armijo(Alpen-Adria-Universität Klagenfurt), Christian Timmerer (Alpen-Adria-Universität Klagenfurt) and Hermann Hellwagner (Alpen-Adria-Universität Klagenfurt)
Abstract: Mobile networks equipped with edge computing nodes enable access to information that can be leveraged to assist client-based adaptive bitrate (ABR) algorithms in making better adaptation decisions to improve both Quality of Experience (QoE) and fairness. For this purpose, we propose a novel on-the-fly edge mechanism, named EADAS (Edge Assisted Adaptation Scheme for HTTP Adaptive Streaming), located at the edge node that assists and improves the ABR decisions on-the-fly. EADAS proposes (i) an edge ABR algorithm to improve QoE and fairness for clients and (ii) a segment prefetching scheme. The results show a QoE increase of 4.6%, 23.5%, and 24.4% and a fairness increase of 11%, 3.4%, and 5.8% when using a buffer-based, a throughput-based, and a hybrid ABR algorithm, respectively, at the client compared with client-based algorithms without EADAS. Moreover, QoE and fairness among clients can be prioritized using parameters of the EADAS algorithm according to service providers’ requirements.
Keywords: Dynamic Adaptive Streaming over HTTP (DASH), Edge Computing, Network-Assisted Video Streaming, Quality of Experience (QoE).
Four papers were presented during this session as shown below:
1. VMAF-based Bitrate Ladder Estimation for Adaptive Streaming
Authors: Angeliki Katsenou (University of Bristol); Fan Zhang (University of Bristol); Kyle Swanson (Netflix); Mariana Afonso (Netflix); Joel Sole (Netflix); David Bull (University of Bristol)
Abstract: In HTTP Adaptive Streaming, video content is conventionally encoded by adapting its spatial resolution and quantization level to best match the prevailing network state and display characteristics. It is well known that the traditional solution, of using a fixed bitrate ladder, does not result in the highest quality of experience for the user. Hence, in this paper, we introduce a content-driven approach for estimating the bitrate ladder, based on spatio-temporal features extracted from the uncompressed content. The method implements a content-driven interpolation. It uses the extracted features to train a
machine learning model to infer the curvature points of the Rate-VMAF curves in order to guide a set of initial encodings. We employ the VMAF quality metric as a means of perceptually conditioning the estimation. When compared to the generation of a reference ladder using exhaustive encoding, 76.63% the estimated ladder’s Rate-VMAF points are identical to those of the reference ladder. The proposed method benefits from a significant
(77.4%) reduction in the number of encodes required with only a small (1.04%) average Bjøntegaard Delta Rate increase.
2. Efficient Multi-Encoding Algorithms for HTTP Adaptive Bitrate Streaming
Authors: Vignesh V Menon (Alpen-Adria-Universitat Klagenfurt); Hadi Amirpour (Alpen-Adria-Universität Klagenfurt); Christian Timmerer (Alpen-Adria-Universität Klagenfurt, Austria); Mohammad Ghanbari (University of Essex, UK)
Abstract: Since video accounts for the majority of today’s internet traffic, the popularity of HTTP Adaptive Streaming (HAS) is increasing steadily. In HAS, each video is encoded at multiple bitrates and spatial resolutions (i.e., representations) to adapt to a heterogeneity of network conditions, device characteristics, and end-user preferences. Most of the streaming services utilize cloud-based encoding techniques which enable a fully parallel encoding process to speed up the encoding and consequently to reduce the overall time complexity. State-of-the-art approaches further improve the encoding process by utilizing encoder analysis information from already encoded representation(s) to improve the encoding time complexity of the remaining representations. In this paper, we investigate various multi-encoding algorithms (i.e., multi-rate and multi-resolution) and propose novel multi-encoding algorithms for large-scale HTTP Adaptive Streaming deployments. Experimental results demonstrate that the proposed multi-encoding algorithm optimized for the highest compression efficiency reduces the overall encoding time by 39% with a 1.5% bitrate increase compared to stand-alone encodings. Its optimized version for the highest time savings reduces the overall encoding time by 50% with a 2.6% bitrate increase compared to standalone encodings.
3. Open GOP Resolution Switching in HTTP Adaptive Streaming with VVC
Authors: Robert Skupin (Fraunhofer HHI); Christian Bartnik (Fraunhofer HHI); Adam Wieckowski (HHI); Yago Sanchez de la Fuente (Fraunhofer HHI); Benjamin Bross (HHI); Cornelius Hellge (Fraunhofer HHI); Thomas Schierl (Fraunhofer HHI)
Abstract: The user experience in adaptive HTTP streaming relies on offering bitrate ladders with suitable operation points for all users and typically involves multiple resolutions. While open GOP coding structures are generally known to provide
substantial coding efficiency benefit, their use in HTTP streaming has been precluded through lacking support of reference picture resampling (RPR) in AVC and HEVC. The
newly emerging Versatile Video Coding (VVC) standard supports RPR, but only conversational scenarios were primarily investigated during the design of VVC. This paper aims at enabling usage of RPR in HTTP streaming scenarios through analysing the drift potential of VVC coding tools and presenting a constrained encoding method that avoids severe drift artefacts in resolution switching with open GOP coding in VVC. In
typical live streaming configurations, the presented method achieves -8.7% BD-rate reduction compared to closed GOP coding while in a typical Video on Demand configuration, -1.89% BD-rate reduction is reported. The constraints penalty
compared to regular open GOP coding is 0.65% BD-rate in the worst case. The presented method was integrated into the publicly available open source VVC encoder VVenC v0.3.
The source code of VVenc can be accessed here. The source code of the VVC reference software (VTM) can be accessed here.
4. Towards Understanding of the Behavior of Web Streaming
Authors: Yuriy Reznik (Brightcove, Inc.); Karl Lillevold (Brightcove, Inc.); Abhijith Jagannath (Brightcove, Inc.); Xiangbo Li (Brightcove, Inc.
Abstract: We study the behavior of a modern-era adaptive streaming system delivering videos embedded in web-pages. In such an application, the size of videos rendered on the screen may depend on user preferences, such as the position and size of a browser window. Moreover, the stream selection logic in such a system appears to be influenced not only by the available network bandwidth but also by the output video size, which, in many cases, limits the selection of higher quality streams. To explain this behavior, in this paper we introduce a simple analytical model of a client adapting to both bandwidth and player size. Using this model, we then compute stream selection probabilities and show
that they are sufficiently close to respective statistics observed in practical experiments. Possible uses of this proposed client model are also suggested. Specifically, we show how it can be used to derive formulae for the average performance parameters of the system and also for posing related optimization problems.
The dataset as mentioned during the presentation can be accessed here.
We thank the organizers of PCS’21, the authors, and reviewers of the presented papers, the attendees who participated with probing questions, making the session successful.
Posted inATHENA|Comments Off on Throwback on our special session at PCS’21
Farzad Tashtarian (AAU, Austria), Abdelhak Bentaleb (National University of Singapore), Reza Farahani(AAU, Austria), Minh Nguyen (AAU, Austria), Christian Timmerer (AAU, Austria), Hermann Hellwagner (AAU, Austria), and Roger Zimmermann (National University of Singapore)
Abstract: Live User Generated Content (UGC) has become very popular in today’s video streaming applications, in particular with gaming and e-sport. However, streaming UGC presents unique challenges for video delivery. When dealing with the technical complexity of managing hundreds or thousands of concurrent streams that are geographically distributed, UGCsystems are forces to made difficult trade-offs with video quality and latency. To bridge this gap, this paper presents a fully distributed architecture for UGC delivery over the Internet, termed QuaLA(joint Quality-Latency Architecture). The proposed architecture aims to jointly optimize video quality and latency for a better user experience and fairness. By using the proximal Jacobi alternating direction method of multipliers(ProxJ-ADMM) technique, QuaLA proposes a fully distributed mechanism to achieve an optimal solution. We demonstrate the effectiveness of the proposed architecture through real-world experiments using the CloudLAB testbed. Experimental results show the outperformance ofQuaLAin achieving high quality with more than 57% improvement while preserving a good level of fairness and respecting a given target latency among all clients compared to conventional client-driven solutions
Keywords: UGC streaming, low latency live streaming, fair-ness, QoE, HAS, DASH, ABR, adaptive streaming, ADMM.
Reza Farahani(Alpen-Adria-Universität Klagenfurt), Farzad Tashtarian (Alpen-Adria-Universität Klagenfurt), Hadi Amirpour (Alpen-Adria-Universität Klagenfurt), Christian Timmerer (Alpen-Adria-Universität Klagenfurt), Mohammad Ghanbari (School of Computer Science and Electronic Engineering, University of Essex, Colchester, UK) and Hermann Hellwagner (Alpen-Adria-Universität Klagenfurt)
Abstract: With the increasing demand for video streaming applications, HTTP Adaptive Streaming (HAS) technology has become the dominant video delivery technique over the Internet. Current HAS solutions only consider either client- or server-side optimization, which causes many problems in achieving high-quality video, leading to sub-optimal users’ experience and network resource utilization. Recent studies have revealed that network-assisted HAS techniques, by providing a comprehensive view of the network, can lead to more significant gains in HAS system performance. In this paper, we leverage the capability of Software-Define Networking (SDN), Network Function Virtualization (NFV), and edge computing to introduce a CDN-Aware QoE Optimization in SDN-Assisted Adaptive Video Streaming framework called CSDN. We employ virtualized edge entities to collect various information items (e.g., user-, client, CDN- and network-level information) in a time-slotted method. These components then run an optimization model with a new server/segment selection approach in a time-slotted fashion to serve the clients’ requests by selecting optimal cache servers (in terms of fetch and transcoding times). In case of a cache miss, a client’s request is served (i) by an optimal replacement quality (only better quality levels with minimum deviation) from a cache server, (ii) by a quality transcoded from an optimal replacement quality at the edge, or (iii) by the originally requested quality level from the origin server. By means of comprehensive experiments conducted on a real-world large-scale testbed, we demonstrate that CSDN outperforms the state-of-the-art in terms of playback bitrate, the number of quality switches, the number of stalls, and bandwidth usage by at least 7.5%, 19%, 19%, and 63%, respectively.
Keywords: Dynamic Adaptive Streaming over HTTP (DASH), Edge Computing, Network-Assisted Video Streaming, Quality of Experience (QoE), Software Defined Networking (SDN), Network Function Virtualization (NFV), Video Transcoding, Content Delivery Network (CDN).
