Zodiac 2 : An Authoring Tool for Active Video Content

Conventional digital video playback systems provide only l imited user interactivity, mostly in the form of VCRlike controls such as fast forward/backward, slow motion, p ause/stop, etc. In this model, the temporal ordering and the spatial viewpoints of the video streams being viewed ar completely determined at authoring time. In contrast, this work describes an Active Videosystem, which supports hyper-linking among related video sequences and interpolation of video sequences with neighb oring viewpoints, to offer end users the additional flexibility of choosing the temporal sequencing and the spat ial viewing angle at playback time, respectively. The proposed active video system uses image-based renderin g (IBR) technology to generate video sequences corresponding to novel viewpoints, and demonstrates that o bject tracking not only enables temporal navigation in the form of “hyper-video,” but also simplifies the applica tion of IBR in authoring active video streams. In this paper we present the design and implementation of Zodiac2, an editing tool specifically designed to support the creation of active video content. The main design goals of Zodiac2are to ease the active video authoring process with object tracking and to reduce additional storage costs and disk bandwidth overhead associated with active video streams using lossy and lossless compression techniq ues.

[1]  Steven M. Seitz,et al.  Image-based transformation of viewpoint and scene appearance , 1997 .

[2]  Antonio Ortega,et al.  Stereo image compression based on disparity field segmentation , 1997, Electronic Imaging.

[3]  Jr. Leonard McMillan,et al.  An Image-Based Approach to Three-Dimensional Computer Graphics , 1997 .

[4]  V. Bove,et al.  Hyperlinked video , 1999, Other Conferences.

[5]  Tzi-cker Chiueh,et al.  Design and Implementation of the Stony Brook Video Server , 1997 .

[6]  Ramesh C. Jain,et al.  An architecture for multiple perspective interactive video , 1995, MULTIMEDIA '95.

[7]  E. Adelson,et al.  The Plenoptic Function and the Elements of Early Vision , 1991 .

[8]  Voicu Popescu,et al.  The WarpEngine: an architecture for the post-polygonal age , 2000, SIGGRAPH.

[9]  Avishai Henik,et al.  Compression of stereo images using subsampling and transform coding , 1991 .

[10]  Sean Hayes,et al.  Automatically Synthesising Virtual Viewpoints by Trinocular Image Interpolation , 1997 .

[11]  Yuh-Lin Chang,et al.  Tracking deformable objects with the active contour model , 1997, Proceedings of IEEE International Conference on Multimedia Computing and Systems.

[12]  Michael Isard,et al.  Active Contours , 2000, Springer London.

[13]  Ian E. Smith,et al.  HyperCafe: narrative and aesthetic properties of hypervideo , 1996, HYPERTEXT '96.

[14]  Takeo Kanade,et al.  Constructing virtual worlds using dense stereo , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[15]  M. Lukacs,et al.  Predictive coding of multi-viewpoint image sets , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[16]  Mubarak Shah,et al.  A fast algorithm for active contours , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[17]  Eran A. Edirisinghe,et al.  Algorithm for compression of stereo image pairs , 1997 .

[18]  Tulika Mitra,et al.  Zodiac: A history-based interactive video authoring system , 1998, MULTIMEDIA '98.

[19]  Yoshiaki Shirai,et al.  Three-Dimensional Computer Vision , 1987, Symbolic Computation.