Abstract
The aim of the present work is to demonstrate the benefit of integrating advanced computer vision techniques into a new generation of compositing tools for the post-production industry. The recent MPEG-4 image description standard supporting the overlay of multiple objects and sprites corresponds closely to the composition process and underpins our wider goal of creating the virtual post-production studio: collaborative creation of multimedia content between studios. Essential to this aim is the creation of efficient and robust compositing tools: in particular, rotoscoping and mosaicking functionality. Rotoscoping involves the separation of elements within a scene: an intensively manual procedure that we automate using active surfaces supported by edge chains. Mosaicking is a new and particularly exciting technique offering the post-production operator the opportunity to easily modify viewing trajectories, introduce computer-generated elements, stabilize jumpy action or positionally justify action in the viewport. Deriving a sophisticated three-dimensional motion model, an optical flow framework is used to facilitate the seamless merging of frames to create a larger virtual scene.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
P Anandan, JR Bergen, KJ Hanna and R Hingorani (1993) Motion analysis and image sequence processing, in Hierarchical Model-Based Motion Estimation (eds. MI Sezan and RL Lagendijk), Kluwer Academic, Boston, pp. 1–22.
S Ayer, P Schroeter and J Bigün (1994) Segmentation of moving objects by robust motion parameter estimation over multiple frame, in Proc. European Conference on Computer Vision, Stockholm, pp. 316–27.
JL Barron, DJ Fleet and SS Beauchemin (1994) Performance of optical flow techniques, International Journal of Computer Vision, 12 (1), 43–77.
M Bober and J Kittler (1994) Robust motion analysis, in Proc. IEEE Computer Soc. Conf on Computer Vision and Pattern Recognition, pp. 947–52.
J Canny (1986) A computational approach to edge detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, 6 (6), 679–98.
L Cohen and I Cohen (1993) finite-element methods for active contour models and balloons for 2D and 3D images, IEEE Transactions on Pattern Analysis and Machine Intelligence, 15 (11), 1l31–47.
JM Corridoni, A Del Bimbo, D Lucarella and WX He (1996) Multi-perspective navigation of movies, Journal of Visual Languages and Computing, 7 (4), 445–466.
D Daneels, D Van Campenhout, W Niblack, W Equitz, R Barber, E Bellon and F Fierens (1993) Interactive outlining: an improved approach using active contours, in Storage and Retrieval for Image and Video Databases, Vol. 1908, pp. 226–33, Proceedings of the International Society of Optical Engineering.
B Duc, P Schroeter and J Bigün (1995) Spatio-temporal robust motion estimation and segmentation, in 6th Int. Conf Computer Analysis of Images and Patterns, Prague, September, Springer-Verlag, p. 238–45.
J Figue (1997) Nemesis project: advanced image analysis tools based on a new 3D wide sense object oriented data representation model for multimedia content creation and post-production applications, in Workshop on Image Analysis for Multimedia Interactive Services, Louvain-Ia-Neuve, Belgium, 24–25 June, pp. 117–26.
E François (1997) Rigid layers reconstruction based on motion segmentation, in Workshop on Image Analysis for Multimedia Interactive Services, Louvain-la-Neuve, Belgium, 24–25 June, pp. 81–6.
M Gelgon and P Bouthemy (1997) A hierarchical motion-based segmentation and tracking technique for video storyboard-like representation and content-based indexing, in Workshop on Image Analysis for Multimedia Interactive Services, Louvain-la-Neuve, Belgium, 24–25 June, pp. 93–8.
PR Giaccone, D Greenhill and GA Jones (1997) Recovering very large visual motion fields, in 10th Scandinavian Conf. on Image Analysis, Finland, June, pp. 917–22.
PR Giaccone and GA Jones (1997) Feed forward estimation of optical flow, in IEE Conf. on Image Processing and its Applications, Dublin, July.
PR Giaccone and GA Jones (1997) Spatio-temporal approaches to the computation of optical flow, in Proc. British Machine Vision Conference, Colchester, UK, September, pp. 420–9.
D Greenhill and GA Jones (1997) Supervised segmentation of cinematographic sequences using active surfaces, in Workshop on Image Analysis for Multimedia Interactive Services, Louvain-la-Neuve, Belgium, 24–25 June, pp. 39–44.
ISO (1994) Information technology - generic coding of moving pictures and associated audio information - Part 2: Video, Dis 13818–2, ISO/IEC.
ISO (1997) Coding of moving pictures and audio: MPEG-7 applications document, N1735, ISO/IEC, Stockholm MPEG Meeting..
GA Jones (1997) Matching corner features using edge contour data, Technical Report KUCSES-97–02, Computer Vision Research Group, Kingston University.
M Kass, A Witkin and D Terzopoulos (1988) Snakes: active contour models, International Journal of Computer Vision, 1, 321–31.
L Kitchen and A Rosenfeld (1982) Graylevel corner detection, PRL, 1, 95–102.
J Kreyss, M Roper, P Alshuth, T Hermes and O Herzog (1997) Video retrieval by still image analysis with ImageMiner, in Proc. International Society for Optical Engineering, Storage and Retrieval for Image and Video Databases, San Jose, 13–14 February, pp. 36–44.
E-P Ong and M Spann (1995) Robust computation of optical flow, in Proc. British Machine Vision Conference, Vol. 2, pp. 573–82, 1995.
M Otte and H-H Nagel (1995) Estimation of optical flow based on higher-order spatiotemporal derivatives in interlaced and non-interlaced image sequences, Artificial Intelligence, 78, 5–43.
J Porril and J Ivins (1994) A semiautomatic tool for 3D medical image analysis using active contour models, Medical Informatics, 19 (1), 81–90.
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag London Limited
About this chapter
Cite this chapter
Giaccone, P.R., Greenhill, D., Jones, G.A. (1999). Semi-Automated Construction of Virtual Scenes from Cinematographic Data. In: Vince, J., Earnshaw, R. (eds) Digital Convergence: The Information Revolution. Springer, London. https://doi.org/10.1007/978-1-4471-0863-4_17
Download citation
DOI: https://doi.org/10.1007/978-1-4471-0863-4_17
Publisher Name: Springer, London
Print ISBN: 978-1-4471-1220-4
Online ISBN: 978-1-4471-0863-4
eBook Packages: Springer Book Archive