In this paper, we propose a new adaptation approach for viewport-adaptive streaming of 360-degree videos over the Internet. The proposed approach is able to systematically decide versions of tiles according to user head movements and network conditions by taking into account not only viewport estimation errors, but also users’ head movements in each segment duration. Experimental results show that the proposed approach can effectively adapt 360-degree videos to both varying network conditions and user head movements. Compared to existing approaches, the proposed approach can improve the average viewport quality by up to 3.8 dB and reduce the standard deviation of the viewport quality by up to 1.1 dB. Also, the impacts of the segment duration and the buffer size are investigated. It is found that long segment durations and large buffer sizes have significant impacts on the performances of tile selection methods.

360-degree video has become increasingly popular nowadays. For effective transmission of bandwidth-intensive 360-degree video over networks, viewport-adaptive streaming has been introduced. In this article, we evaluate, for the first time, ten existing methods to understand the effectiveness of tile-based viewport adaptive streaming of 360-degree video. Experimental results show that tile-based methods can improve the average V-PSNR by up to 4.3 dB compared to a non-tiled method under low delay settings. Here, the V-PSNR is computed as the peak signal-to-noise ratio of the adapted viewport compared to the corresponding origin viewport. Also, different methods show different tradeoffs between average viewport quality and viewport quality variations. Especially, the performances of most tile-based methods decrease quickly as the segment duration and/or buffer size increase for the content with no main focus. Even, under long delay settings like HTTP Adaptive Streaming, it is found that the simple non-tiled method appears to be the best one. For the content with a strong viewing focus, it is found that the tile-based methods are less influenced by the segment duration and the buffer size. In addition, a comparison of the performances of the tile selection methods using two popular viewport estimation methods is conducted. It is interesting that there is only little difference found in performances of tile selection methods. The findings of this study are useful for service providers to make decisions on deployment of streaming solutions.

HTTP Adaptive Streaming has become a popular solution for multimedia delivery nowadays. However, due to network bandwidth fluctuations, video quality strongly varies during streaming. Therefore, a key challenge in HTTP Adaptive Streaming is how to evaluate the overall quality of a streaming session. In this paper, a machine learning approach is proposed for overall quality prediction, where each segment in a streaming session is represented by a set of features. Two options of the feature set are investigated. In the first option, we use four features, namely segment quality, content characteristics, stalling duration, and padding. The second option consists of three features, namely bitstream-level parameters, stalling duration, and padding. The features are fed into a Long Short Term Memory (LSTM) network that is capable of exploring temporal relations between impairment events of quality variations and stalling events. The overall quality is predicted from the outputs of the LSTM network using a linear regression module. Through experimental results, it is shown that the proposed approach achieves a high prediction performance and outperforms seven existing approaches. Especially, the second option is found to be both efficient and effective. The source code of the proposed approach has been made available to the public.

Short videos have become one of the most popular content mobile users consume nowadays. However, unlike traditional videos, users watch many short videos each time and frequently skip those not of their interest. Not taking into account this factor, conventional video streaming methods result in a large amount of data wastage. Moreover, short videos are usually streamed over time-varying network conditions, which requires preloading to provide satisfactory Quality of Experience (QoE). In this paper, we formulate the QoE-Wastage trade-off optimization problem in short video streaming as a multi-objective optimization problem. We then develop a lightweight joint preloading and bitrate adaptation algorithm that utilizes cross-user behaviors to decide a suitable amount of preloaded data for each video. In addition, our method dynamically adjusts video bitrate per chunk basis to achieve high user QoE. Experimental results show that the proposed method improves QoE by approximately 12% to 45% and reduces data wastage by up to 58% compared to existing methods.

Virtual Reality (VR), which brings immersive experiences to viewers, has been gaining popularity in recent years. A key feature in VR systems is the use of omnidirectional content, which provides 360-degree views of scenes. In this work, we study the human quality perception of omnidirectional images, focusing on different zones surrounding the foveation point. For that purpose, an extensive subjective experiment is carried out to assess the perceptual quality of omnidirectional images with non-uniformquality. Through experimental results, the impacts of different zones are analyzed. Moreover, twenty-five objective quality metrics, including foveal quality metrics, are evaluated using our database. It is quantitatively shown that the zones corresponding to the fovea and parafovea of human eyes are extremely important for quality perception, while the impacts of the other zones corresponding to the perifovea and periphery are small. Besides, most of the investigated metrics are found to be not effective enough to reflect the quality perceived by viewers. Our database has been made available to the public.