DAVE: a system for quality driven adaptive video delivery

Delivering relevant video segments (video highlights) to a variety of devices operating under both wired and wireless platforms requires: (a) a mechanism for describing video content, and the environment in which it is to be delivered, and (b) a framework to support physical and semantic adaptation of a video. The aim of the adaptation process is to maximize users' viewing experiences of video quality and minimize resources required to deliver the content. Towards this, we have developed a quality driven adaptation algorithm and implemented a prototype system called DAVE. DAVE minimizes resource requirements by identifying relevant segments within a video and then scales the identified video segment in different dimensions so that it can be delivered within the available network bandwidth, and played back by the client device. Simultaneously, it maximizes the video quality by identifying and tuning relationships between perceptual quality parameters such as frame rate and frame size, with network and device resources such as bit rate, to maximize user-viewing experience. In this paper, we present a quality-driven adaptation algorithm and its implementation through the DAVE system.

[1]  Abdelsalam Helal,et al.  Adaptive delivery of video data over wireless and mobile environments , 2003, Wirel. Commun. Mob. Comput..

[2]  Surya Nepal,et al.  Adaptive video highlights for wired and wireless platforms , 2003, 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698).

[3]  George Lawton,et al.  New Technologies Place Video in Your Hand , 2001, Computer.

[4]  Allan Kuchinsky,et al.  Framework for adaptive content delivery in heterogeneous network environments , 1999, Electronic Imaging.

[5]  Zhijun Lei.  Media transcoding for pervasive computing , 2001, MULTIMEDIA '01.

[6]  John R. Smith,et al.  Adapting Multimedia Internet Content for Universal Access , 1999, IEEE Trans. Multim..

[7]  John R. Smith,et al.  Scalable multimedia delivery for pervasive computing , 1999, MULTIMEDIA '99.

[8]  Chung-Ming Huang,et al.  A proxy-based adaptive flow control scheme for media streaming , 2002, SAC '02.

[9]  Julie A. McCann,et al.  Kendra: adaptive Internet system , 2000, J. Syst. Softw..

[10]  C. van den Branden Lambrecht.  A working spatio-temporal model of the human visual system for image restoration and quality assessment applications , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[11]  Nicolas D. Georganas,et al.  Context-based media adaptation in pervasive computing , 2001, Canadian Conference on Electrical and Computer Engineering 2001. Conference Proceedings (Cat. No.01TH8555).

[12]  Christian J. Van Den Branden Lambrecht.  A working spatio-temporal model of the human visual system for image restoration and quality assessment applications , 1996, ICASSP.

[13]  Ranga S. Ramanujan,et al.  Adaptive streaming of MPEG video over IP networks , 1997, Proceedings of 22nd Annual Conference on Local Computer Networks.

[14]  Thomas D. C. Little,et al.  The Use of Metadata for the Rendering of Personalized Video Delivery , 1998, Multimedia Data Management.

[15]  Deborah Estrin,et al.  Layered quality adaptation for Internet video streaming , 2000, IEEE Journal on Selected Areas in Communications.

[16]  Surya Nepal,et al.  Automatic detection of 'Goal' segments in basketball videos , 2001, MULTIMEDIA '01.