PV '18- Proceedings of the 23rd Packet Video Workshop

Originally proposed by Google, QUIC is a low-latency transport protocol currently being developed and specified in the IETF. QUIC's low-latency, improved congestion control, multiplexing features are promising and may help improve viewer experience in HTTP adaptive streaming applications. To investigate what issues due to running HTTP over TCP can be alleviated by using HTTP over QUIC, we measured QUIC's streaming performance on wireless and cellular networks. Specifically, we examined QUIC's performance during network interface changes due to viewer's mobility and under unstable network conditions. Results show that QUIC starts media streams more quickly, providing a better streaming and seeking experience, in particular, when there is more congestion in the network, and outperforms TCP when the viewer is mobile and switches between the networks.

Video Quality assessment is imperative to estimate and hence manage the Quality of Experience (QoE) in video streaming applications to the end-user. Recent years have seen a tremendous advancement in the field of objective video quality assessment (VQA) metrics, with the development of models that can predict the quality of the videos streamed over the Internet. However, no work so far has attempted to study the performance of such quality assessment metrics on gaming videos, which are artificial and synthetic and have different streaming requirements than traditionally streamed videos. Towards this end, we present in this paper a study of the performance of objective quality assessment metrics for gaming videos considering passive streaming applications. Objective quality assessment considering eight widely used VQA metrics is performed on a dataset of 24 reference videos and 576 compressed sequences obtained by encoding them at 24 different resolution-bitrate pairs. We present an evaluation of the performance behavior of the VQA metrics. Our results indicate that VMAF predicts subjective video quality ratings the best, while NIQE turns out to be a promising alternative as a no-reference metric in some scenarios.

Adapting the transmission rate of video telephony Internet applications in order to guarantee the maximal communication quality is still an open and extremely challenging problem. The congestion control algorithm, which is the algorithm responsible for adjusting the transmission rate according to the network conditions, should typically be able to reach the largest possible rate, in order to achieve a high video quality, at the minimum possible delay, in order to guarantee a good interactivity. At the same time, it should also guarantee a fair share of the network resources when competing with other communication protocols, in particular loss-based congestion protocols. These two objectives actually conflict with each other: whereas, in order to achieve the largest rate with the minimum delay, the delay-based congestion control should be extremely sensitive to delay variations, it should also be ideally immune to delay variations to have perfect coexistence with loss-based protocols. In order to achieve this double objective we propose a learning-based adaptive controller that tunes the delay sensitivity of an underlying delay-based congestion control according to the estimated network conditions. We first define a simple low-dimensional model for the network response. We then formulate a bayesian bandit problem for the selection of the delay sensitivity of the congestion control algorithm. By solving the bandit problem using an optimal learning method we are able to maximize effectively the long term utility provided to the user. Finally, we provide simulation results to demonstrate the operation of the proposed method and its effective ability to adapt to different network scenarios in order to maximize the communication quality.

This paper analyzes and improves an existing guided transcoding scheme called deflation and inflation. By adding three new coding tools, required storage is reduced by 4.2 percentage points on a test set defined by the Moving Picture Experts Group. The first tool predicts coefficient signs, the second tool reorders coefficients using a dynamic scanning order, and the third tool uses new contexts for the binary coder to encode whether coefficients are non-zero or not. All three tools are based on a new method for estimating the likelihood of a coefficient being non-zero. Furthermore, the paper introduces a modified test configuration using a fast transcoder based on the x265 encoder for comparisons. Using this configuration, the three proposed tools provide a storage reduction of 4.4 percentage points with a transcoding complexity of around 5% of a transcoder based on x265.

High-quality point clouds have recently gained interest as an emerging form of representing immersive 3D graphics. Unfortunately, these 3D media are bulky and severely bandwidth intensive, which makes it difficult for streaming to resource-limited and mobile devices. This has called researchers to propose efficient and adaptive approaches for streaming of high-quality point clouds.

In this paper, we run a pilot study towards dynamic adaptive point cloud streaming, and extend the concept of dynamic adaptive streaming over HTTP (DASH) towards DASH-PC, a dynamic adaptive bandwidth-efficient and view-aware point cloud streaming system. DASH-PC can tackle the huge bandwidth demands of dense point cloud streaming while at the same time can semantically link to human visual acuity to maintain high visual quality when needed. In order to describe the various quality representations, we propose multiple thinning approaches to spatially sub-sample point clouds in the 3D space, and design a DASH Media Presentation Description manifest speci.c for point cloud streaming. Our initial evaluations show that we can achieve signi.cant bandwidth and performance improvement on dense point cloud streaming with minor negative quality impacts compared to the baseline scenario when no adaptations is applied.

Video Streaming service providers are dealing with increasingly large media asset content repositories. To store these in a flexible and persistent manner, cloud storage is often used. In addition the compute power available in the cloud is often used for on-the-fly format conversion such as dynamic packaging. This way, additional protocols, encryption schemes and formats can be supported from a single storage source. We present a study on the performance of this increasingly common video streaming setup and propose an improvement. We compare the efficiency of different storage formats used in the cloud storage such as fragmented MPEG-4 and MPEG-4 and compare local versus cloud based object storage. In addition we study the resulting traffic between the storage, compute and client nodes. Results show reduced traffic towards the client in the cloud storage based solution. Resulting traffic patterns hint that specific parts of the media assets are most relevant for the conversion at the compute node. Based on this we then introduce an improvement scheme that involves caching metadata files based on dynamically referenced MPEG-4 at the compute node. Experimental results show many benefits such as halving the request time (-50 percent for segment request, -90 percent for manifest requests), reduced number of calls to storage and improved throughput towards the client (+20 percent for fMP4, +100 percent for MP4). In addition, experiments showed player playback startup delay reduced up to 60%. Contrary to the initial setup, both MP4 and fMP4 based storage result in good streaming performance with this scheme, allowing both formats to be used in the storage backend.

Streaming the whole omnidirectional video at high quality is less efficient in terms of bandwidth requirements and decoding complexity, since only a small part of the 360-degree horizontal field of view can be seen by a user at a given point in time. In Viewport Dependent Streaming (VDS) only the current user viewport is streamed at high quality, while the remaining parts are streamed at lower quality. This technology may save streaming bandwidth considerably, especially when it is associated to other techniques. Among the others, asymmetric stereoscopic video, has been studied in the past for traditional video and displays. We focused our research on the usage of asymmetric stereoscopic video for omnidirectional streams watched with a Head Mounted Display (HMD) in VDS. We conducted two subjective quality experiments with the main goal of reducing the streaming bandwidth, while keeping the subjective video quality at the highest level. We assessed asymmetric video applied separately to the foreground and background views of omnidirectional VDS sessions. We show that for VDS, applying asymmetric stereoscopic streaming delivery on the foreground view can save up to 41% bit rate, and using the same technique on the background view can save approximately up to 15% bit rate. Furthermore, eye dominance was seen not to be relevant in our experiments.

We discuss the problem of optimal design of encoding profiles for adaptive bitrate (ABR) streaming. We formalize this problem and show that it belongs to a class of non-linear constrained optimization problems, with several methods available for solving it numerically. We illustrate the effectiveness of our approach by several examples of optimal encoding ladders constructed for different sources and network models.

In-loop filtering is an important task in video coding, as it refines both the reconstructed signal for display and the pictures used for inter-prediction. In order to remove coding artifacts, machine learning based methods are assumed to be beneficial, as they utilize some prior knowledge on the characteristics of raw images. In this contribution, a dictionary learning / sparse coding based inloop filter and a frequency adaptation model based on the lp-ballenergy in the spectral domain is proposed. Thereby the dictionary is trained on raw data and the algorithms are controlled mainly by the parameter for the sparsity. The frequency adaption model results in further improvement of the sparse coding based loop filter. Experimental results show that the proposed method results in coding gains up to l-4.6 % at peak and -1.74 % on average against HEVC in a Random Access coding configuration.

Delivering video content with a high and fairly shared quality of experience is a challenging task in view of the drastic video traffic increase forecasts. Currently, content delivery networks provide numerous servers hosting replicas of the video content, and consuming clients are re-directed to the closest server. Then, the video content is streamed using adaptive streaming solutions. However, some servers become overloaded, and clients may experience a poor or unfairly distributed quality of experience.

In this paper we propose Muslin, a streaming solution supporting a high, fairly shared end-users quality of experience for live streaming. Muslin leverages on MS-Stream, a content delivery solution in which a client can simultaneously use several servers. Muslin dynamically provisions servers and replicates content into servers, and advertises servers to clients based on real-time delivery conditions. We have used Muslin to replay a one-day video-games event, with hundreds of clients and several test beds. Our results shows that our approach outperforms traditional content delivery schemes by increasing the fairness and quality of experience at the user side without requiring a greater underlying content delivery platform.

About 300 hours of video are uploaded to YouTube every minute. The main technology to delivery YouTube content to various clients is HTTP adaptive streaming and the majority of today's internet traffic comprises streaming audio and video. In this paper, we investigate content provisioning for HTTP adaptive streaming under predefined aspects representing content features and upload characteristics as well and apply it to YouTube. Additionally, we compare the YouTube's content upload and processing functions with a commercially available video encoding service. The results reveal insights into YouTube's content upload and processing functions and the methodology can be applied to similar services. All experiments conducted within the paper allow for reproducibility thanks to the usage of open source tools, publicly available datasets, and scripts used to conduct the experiments on virtual machines.

The efficient provision of IPTV services requires support for IP multicasting and IGMP snooping, limiting such services to single operator networks. Information-Centric Networking (ICN), with its native support for multicast seems ideal for such services, but it requires operators and users to overhaul their networks and applications. The POINT project has proposed a hybrid, IP-over-ICN, architecture, preserving IP devices and applications at the edge, but interconnecting them via an SDN-based ICN core. This allows individual operators to exploit the benefits of ICN, without expecting the rest of the Internet to change. In this paper, we first outline the POINT approach and show how it can handle multicast-based IPTV services in a more efficient and resilient manner than IP. We then describe a successful trial of the POINT prototype in a production network, where real users tested actual IPTV services over both IP and POINT under regular and exceptional conditions. Results from the trial show that the POINT prototype matched or improved upon the services offered via plain IP.

The recent Omnidirectional MediA Format (OMAF) standard specifies delivery of 360° video content. OMAF supports only equirectangular (ERP) and cubemap projections and their region-wise packing with a limitation on video decoding capability to the maximum resolution of 4K (e.g., 4096x2048). Streaming of 4K ERP content allows only a limited viewport resolution, which is lower than the resolution of many current head-mounted displays (HMDs). In order to take the full advantage of those HMDs, this work proposes a specific mixed-resolution packing of 6K (6144x3072) ERP content and its realization in tile-based streaming, while complying with the 4K-decoding constraint and the High Efficiency Video Coding (HEVC) standard. Experimental results indicate that, using Zonal-PSNR test methodology, the proposed layout decreases the streaming bitrate up to 32% in terms of BD-rate, when compared to mixed-quality viewport-adaptive streaming of 4K ERP as an alternative solution.