Modeling of Rate and Perceptual Quality of Compressed Video as Functions of Frame Rate and Quantization Stepsize and Its Applications

Ma, Zhan; Xu, Meng; Ou, Yen-Fu; Wang, Yao

doi:10.1109/tcsvt.2011.2177143

Cited by 97 publications

(100 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Since talking head videos are used in this study and the VQD system is dealing with the videos frame by frame before passing the frames to the codec, the value of b was approximately 1. However, according to other studies with diverse video content, b has been found to vary with the intensity of motion [17]. The parameter d had the value of 0.6312.…”

Section: Bit-rate Modelmentioning

confidence: 84%

Minimisation of video downstream bit rate for large scale immersive video conferencing by utilising the perceptual variations of quality

Pourashraf

Safaei

Franklin

2014

2014 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

This paper aims at minimising the video downstream bit rate of immersive video conferencing (IVC) applications by judiciously modifying the video quality based on the relative virtual positions of participants in the virtual environment. The paper reports on the results of a user study to assess the influence of participants' perspectives on the perceptual impact of relevant video parameters, such as resolution and frame rate. A mathematical model for video rate is proposed that expresses the total rate as the product of spatial resolution and frame rate. Results from the user study are combined with the proposed model to predict the rate parameters which will result in perceptually acceptable quality for a given user perspective. The simulation results show that by exploiting the proposed method, the downstream network load for the client can be significantly reduced with little or no impact on the perceived quality. ABSTRACTThis paper aims at minimising the video downstream bit rate of immersive video conferencing (IVC) applications by judiciously modifying the video quality based on the relative virtual positions of participants in the virtual environment. The paper reports on the results of a user study to assess the influence of participants' perspectives on the perceptual impact of relevant video parameters, such as resolution and frame rate. A mathematical model for video rate is proposed that expresses the total rate as the product of spatial resolution and frame rate. Results from the user study are combined with the proposed model to predict the rate parameters which will result in perceptually acceptable quality for a given user perspective. The simulation results show that by exploiting the proposed method, the downstream network load for the client can be significantly reduced with little or no impact on the perceived quality.Index Terms-Video quality differentiation, video quality assessments, video conferencing, 3D immersive environments, rate model, perceptual model

show abstract

Section: Bit-rate Modelmentioning

confidence: 84%

Minimisation of video downstream bit rate for large scale immersive video conferencing by utilising the perceptual variations of quality

Pourashraf

Safaei

Franklin

2014

2014 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

show abstract

“…For the parametric modeling of MOS (i.e., video quality or QoE) with respect to SVC scalability (t and QP ), following [35], [36], we define perceptive video quality, Q(q, t) that approximates the MOS, where q = 2 (QP −4)/6 . Specifically, as noted in [35], the parametric quality measure has a direct relationship with the subjective measure MOS, given by equation (7): MOS = 4 × Q(q, t) + 1…”

Section: B Subjective Video Quality Assessment Results and Parametrimentioning

confidence: 99%

eWU-TV: User-Centric Energy-Efficient Digital TV Broadcast Over Wi-Fi Networks

Singhal

Trestian

et al. 2015

IEEE Trans. on Broadcast.

View full text Add to dashboard Cite

Abstract-This paper presents an innovative multi-faceted architecture, named eWU-TV, that provides an energy efficient, user-centric, adaptive digital television (DTV) broadcast over WiFi. To cater to the varied properties of the user equipments (UEs), the proposed framework broadcasts DTV content in the form of scalable video coded content that is adapted to suit the subscribers' requirements. The user-centricity is in terms of UE device display size, user preferences for video quality profile based on device energy saving, and UE transmission technology support (DVB-T/H or Wi-Fi). Mathematical models on device battery discharge, QoE, and user preference are devised that closely approximate the results of device battery discharge experiments on DTV reception by heterogeneous devices over Wi-Fi/DVB-T, subjective video quality assessment study, and statistical survey of user preference. The proposed eWU-TV performance optimization framework is based on the developed models. The framework ensures that the adaptive scalable broadcast reception via Wi-Fi serves more number of users with higher quality of user experience and with provisions for significant device energy saving.

show abstract

“…To this end, in the following, we introduce the Quality of Experience (QoE) as an additional requirement in the eMBMS system design. We use recently proposed QoE models for H.264/SVC services that provide analytical relations between the average data rates, R(q, t), and average subjective video quality based on Mean Opinion Score (MOS), Q(q, t), as a function of major H.264/SVC parameters: the frame rate (t) (Hz) and the quantization stepsize parameter (q) [30]. In this paper, we choose the rate and quality model parameters and the encoder settings, that are provided in [30] for the Foreman video sequence encoded using the SVC reference software [32].…”

Section: Qoe-aware Embms Designmentioning

confidence: 99%

“…We use recently proposed QoE models for H.264/SVC services that provide analytical relations between the average data rates, R(q, t), and average subjective video quality based on Mean Opinion Score (MOS), Q(q, t), as a function of major H.264/SVC parameters: the frame rate (t) (Hz) and the quantization stepsize parameter (q) [30]. In this paper, we choose the rate and quality model parameters and the encoder settings, that are provided in [30] for the Foreman video sequence encoded using the SVC reference software [32]. Using the QoE models, we reformulate the eMBMS design problem and analyze the bandwidth and energy efficiency per unit of service in a way which is natural for mobile network operators: What are the bandwidth/energy costs per unit of eMBMS service satisfying certain average video quality threshold Q th for a given eMBMS system configuration?…”

Section: Qoe-aware Embms Designmentioning

confidence: 99%

Energy Efficient Multimedia Delivery Services over LTE/LTE-A

Khirallah

Rastovac

Vukobratović

et al. 2014

IEICE Trans. Commun.

View full text Add to dashboard Cite

SUMMARYMobile video services are becoming a dominant traffic category in emerging fourth generation (4G) cellular networks such as the 3GPP Long-Term Evolution (LTE) and LTE-Advanced (LTE-A). In particular, mobile video unicasting services based on 3GPP Dynamic Adaptive Streaming over HTTP (DASH) and multicasting/broadcasting services based on 3GPP evolved Multimedia Multicast/Broadcast Service (eM-BMS) will require considerable resources for high-quality service delivery with high coverage probability. Faced with the challenge of energy efficient multimedia service provisioning over LTE/LTE-A, in this paper, we present simple analytical tools for evaluation of average service data rates, bandwidth and energy-consumption requirements applicable for different multimedia delivery services and LTE/LTE-A radio access network (RAN) configurations. Moreover, we introduce and evaluate novel energy and bandwidth performance measures defined per unit of service. As a result, we are able to compare the efficiency of different multimedia service delivery configurations over LTE/LTE-A. In particular, in this paper, as a running example we focus on eMBMS and compare the Energy of Service (EoS) of the two macro-cellular LTE/LTE-A configurations recently proposed in 3GPP: i) a single frequency network eMBMS (SFN-eMBMS), and ii) a single-cell eMBMS (SC-eMBMS). Furthermore, we extend this analysis to eMBMS provisioning over Heterogeneous Networks (HetNets) environment. However, the methodology presented is general and targets light-weight system design and comparison of bandwidth/energy costs of different LTE/LTE-A multimedia service delivery configurations. key words: eMBMS, energy efficiency, video multicast, LTE/LTE-A

show abstract

Modeling of Rate and Perceptual Quality of Compressed Video as Functions of Frame Rate and Quantization Stepsize and Its Applications

Cited by 97 publications

References 9 publications

Minimisation of video downstream bit rate for large scale immersive video conferencing by utilising the perceptual variations of quality

Minimisation of video downstream bit rate for large scale immersive video conferencing by utilising the perceptual variations of quality

eWU-TV: User-Centric Energy-Efficient Digital TV Broadcast Over Wi-Fi Networks

Energy Efficient Multimedia Delivery Services over LTE/LTE-A

Contact Info

Product

Resources

About