Extraction of Semantic Dynamic Content from
Videos with Probabilistic Motion Models 1st
edition by Gwenaelle Piriou, Patrick Bouthemy,
Jian Feng Yao ISBN 3540219828 9783540219828 pdf
download
https://ebookball.com/product/extraction-of-semantic-dynamic-
content-from-videos-with-probabilistic-motion-models-1st-edition-
by-gwenaelle-piriou-patrick-bouthemy-jian-feng-yao-
isbn-3540219828-9783540219828-9656/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
Tracking Articulated Motion Using a Mixture of Autoregressive Models
1st edition by Ankur Agarwal, Bill Triggs ISBN 3540219828
9783540219828
https://ebookball.com/product/tracking-articulated-motion-using-
a-mixture-of-autoregressive-models-1st-edition-by-ankur-agarwal-
bill-triggs-isbn-3540219828-9783540219828-13696/
Braving the Semantic Gap Mapping Visual Concepts from Images and
Videos 1st Edition by Da Deng 9783540240549
https://ebookball.com/product/braving-the-semantic-gap-mapping-
visual-concepts-from-images-and-videos-1st-edition-by-da-
deng-9783540240549-11540/
Joint Bayes Filter A Hybrid Tracker for Non-rigid Hand Motion
Recognition 1st edition by Huang Fei, Ian Reid ISBN 3540219828
9783540219828
https://ebookball.com/product/joint-bayes-filter-a-hybrid-
tracker-for-non-rigid-hand-motion-recognition-1st-edition-by-
huang-fei-ian-reid-isbn-3540219828-9783540219828-9234/
Real Time Tracking of Multiple Skin Colored Objects with a Possibly
Moving Camera 1st edition by Antonis Argyros, Manolis Lourakis ISBN
3540219828 9783540219828
https://ebookball.com/product/real-time-tracking-of-multiple-
skin-colored-objects-with-a-possibly-moving-camera-1st-edition-
by-antonis-argyros-manolis-lourakis-
isbn-3540219828-9783540219828-13610/
,Learning to Segment 1st edition by Eran Borenstein, Shimon Ullman ISBN
3540219828 9783540219828
https://ebookball.com/product/learning-to-segment-1st-edition-by-
eran-borenstein-shimon-ullman-
isbn-3540219828-9783540219828-10662/
Bias in Shape Estimation 1st edition by Hui Ji, Cornelia Fermuller
ISBN 3540219828 9783540219828
https://ebookball.com/product/bias-in-shape-estimation-1st-
edition-by-hui-ji-cornelia-fermuller-
isbn-3540219828-9783540219828-9210/
Business Economics and Finance with MATLAB GIS and Simulation Model
1st Edition by Patrick L Anderson ISBN 0471719297 9781584883487
https://ebookball.com/product/business-economics-and-finance-
with-matlab-gis-and-simulation-models-1st-edition-by-patrick-l-
anderson-isbn-0471719297-9781584883487-13842/
Intrinsic Images by Entropy Minimization 1st edition by Graham
Finlayson, Mark Drew, Cheng Lu ISBN 3540219828 9783540219828
https://ebookball.com/product/intrinsic-images-by-entropy-
minimization-1st-edition-by-graham-finlayson-mark-drew-cheng-lu-
isbn-3540219828-9783540219828-9776/
Image Similarity Using Mutual Information of Regions 1st edition by
Daniel Russakoff, Carlo Tomasi, Torsten Rohlfing, Calvin Maurer ISBN
3540219828 9783540219828
https://ebookball.com/product/image-similarity-using-mutual-
information-of-regions-1st-edition-by-daniel-russakoff-carlo-
tomasi-torsten-rohlfing-calvin-maurer-
isbn-3540219828-9783540219828-11452/
, Extraction of Semantic Dynamic Content from
Videos with Probabilistic Motion Models
Gwenaëlle Piriou1 , Patrick Bouthemy1 , and Jian-Feng Yao1,2
1
IRISA/INRIA,
2
IRMAR,
Campus universitaire de Beaulieu, 35042 Rennes cedex, France
{Gwenaelle.Piriou,Patrick.Bouthemy,Jian-Feng.Yao}@irisa.fr
Abstract. The exploitation of video data requires to extract informa-
tion at a rather semantic level, and then, methods able to infer “con-
cepts” from low-level video features. We adopt a statistical approach
and we focus on motion information. Because of the diversity of dy-
namic video content (even for a given type of events), we have to design
appropriate motion models and learn them from videos. We have de-
fined original and parsimonious probabilistic motion models, both for
the dominant image motion (camera motion) and the residual image
motion (scene motion). These models are learnt off-line. Motion mea-
surements include affine motion models to capture the camera motion,
and local motion features for scene motion. The two-step event detection
scheme consists in pre-selecting the video segments of potential interest,
and then in recognizing the specified events among the pre-selected seg-
ments, the recognition being stated as a classification problem. We report
accurate results on several sports videos.
1 Introduction and Related Work
Exploiting the tremendous amount of multimedia data, and specifically video
data, requires to develop methods able to extract information at a rather seman-
tic level. Video summarization, video retrieval or video surveillance are examples
of applications. Inferring concepts from low-level video features is a highly chal-
lenging problem. The characteristics of a semantic event have to be expressed
in terms of video primitives (color, texture, motion, shape ...) sufficiently dis-
criminant w.r.t. content. This remains an open problem at the source of active
research activities.
In [9], statistical models for components of the video structure are introduced
to classify video sequences into different genres. The analysis of image motion
is widely exploited for the segmentation of videos into meaningful units or for
event recognition. Efficient motion characterization can be derived from the op-
tical flow, as in [8] for human action change detection. In [11], the authors use
very simple local spatio-temporal measurements, i.e., histograms of the spatial
and temporal intensity gradients, to cluster temporal dynamic events. In [10], a
principal component representation of activity parameters (such as translation,
T. Pajdla and J. Matas (Eds.): ECCV 2004, LNCS 3023, pp. 145–157, 2004.
c Springer-Verlag Berlin Heidelberg 2004
Videos with Probabilistic Motion Models 1st
edition by Gwenaelle Piriou, Patrick Bouthemy,
Jian Feng Yao ISBN 3540219828 9783540219828 pdf
download
https://ebookball.com/product/extraction-of-semantic-dynamic-
content-from-videos-with-probabilistic-motion-models-1st-edition-
by-gwenaelle-piriou-patrick-bouthemy-jian-feng-yao-
isbn-3540219828-9783540219828-9656/
Explore and download more ebooks or textbooks
at ebookball.com
, Get Your Digital Files Instantly: PDF, ePub, MOBI and More
Quick Digital Downloads: PDF, ePub, MOBI and Other Formats
Tracking Articulated Motion Using a Mixture of Autoregressive Models
1st edition by Ankur Agarwal, Bill Triggs ISBN 3540219828
9783540219828
https://ebookball.com/product/tracking-articulated-motion-using-
a-mixture-of-autoregressive-models-1st-edition-by-ankur-agarwal-
bill-triggs-isbn-3540219828-9783540219828-13696/
Braving the Semantic Gap Mapping Visual Concepts from Images and
Videos 1st Edition by Da Deng 9783540240549
https://ebookball.com/product/braving-the-semantic-gap-mapping-
visual-concepts-from-images-and-videos-1st-edition-by-da-
deng-9783540240549-11540/
Joint Bayes Filter A Hybrid Tracker for Non-rigid Hand Motion
Recognition 1st edition by Huang Fei, Ian Reid ISBN 3540219828
9783540219828
https://ebookball.com/product/joint-bayes-filter-a-hybrid-
tracker-for-non-rigid-hand-motion-recognition-1st-edition-by-
huang-fei-ian-reid-isbn-3540219828-9783540219828-9234/
Real Time Tracking of Multiple Skin Colored Objects with a Possibly
Moving Camera 1st edition by Antonis Argyros, Manolis Lourakis ISBN
3540219828 9783540219828
https://ebookball.com/product/real-time-tracking-of-multiple-
skin-colored-objects-with-a-possibly-moving-camera-1st-edition-
by-antonis-argyros-manolis-lourakis-
isbn-3540219828-9783540219828-13610/
,Learning to Segment 1st edition by Eran Borenstein, Shimon Ullman ISBN
3540219828 9783540219828
https://ebookball.com/product/learning-to-segment-1st-edition-by-
eran-borenstein-shimon-ullman-
isbn-3540219828-9783540219828-10662/
Bias in Shape Estimation 1st edition by Hui Ji, Cornelia Fermuller
ISBN 3540219828 9783540219828
https://ebookball.com/product/bias-in-shape-estimation-1st-
edition-by-hui-ji-cornelia-fermuller-
isbn-3540219828-9783540219828-9210/
Business Economics and Finance with MATLAB GIS and Simulation Model
1st Edition by Patrick L Anderson ISBN 0471719297 9781584883487
https://ebookball.com/product/business-economics-and-finance-
with-matlab-gis-and-simulation-models-1st-edition-by-patrick-l-
anderson-isbn-0471719297-9781584883487-13842/
Intrinsic Images by Entropy Minimization 1st edition by Graham
Finlayson, Mark Drew, Cheng Lu ISBN 3540219828 9783540219828
https://ebookball.com/product/intrinsic-images-by-entropy-
minimization-1st-edition-by-graham-finlayson-mark-drew-cheng-lu-
isbn-3540219828-9783540219828-9776/
Image Similarity Using Mutual Information of Regions 1st edition by
Daniel Russakoff, Carlo Tomasi, Torsten Rohlfing, Calvin Maurer ISBN
3540219828 9783540219828
https://ebookball.com/product/image-similarity-using-mutual-
information-of-regions-1st-edition-by-daniel-russakoff-carlo-
tomasi-torsten-rohlfing-calvin-maurer-
isbn-3540219828-9783540219828-11452/
, Extraction of Semantic Dynamic Content from
Videos with Probabilistic Motion Models
Gwenaëlle Piriou1 , Patrick Bouthemy1 , and Jian-Feng Yao1,2
1
IRISA/INRIA,
2
IRMAR,
Campus universitaire de Beaulieu, 35042 Rennes cedex, France
{Gwenaelle.Piriou,Patrick.Bouthemy,Jian-Feng.Yao}@irisa.fr
Abstract. The exploitation of video data requires to extract informa-
tion at a rather semantic level, and then, methods able to infer “con-
cepts” from low-level video features. We adopt a statistical approach
and we focus on motion information. Because of the diversity of dy-
namic video content (even for a given type of events), we have to design
appropriate motion models and learn them from videos. We have de-
fined original and parsimonious probabilistic motion models, both for
the dominant image motion (camera motion) and the residual image
motion (scene motion). These models are learnt off-line. Motion mea-
surements include affine motion models to capture the camera motion,
and local motion features for scene motion. The two-step event detection
scheme consists in pre-selecting the video segments of potential interest,
and then in recognizing the specified events among the pre-selected seg-
ments, the recognition being stated as a classification problem. We report
accurate results on several sports videos.
1 Introduction and Related Work
Exploiting the tremendous amount of multimedia data, and specifically video
data, requires to develop methods able to extract information at a rather seman-
tic level. Video summarization, video retrieval or video surveillance are examples
of applications. Inferring concepts from low-level video features is a highly chal-
lenging problem. The characteristics of a semantic event have to be expressed
in terms of video primitives (color, texture, motion, shape ...) sufficiently dis-
criminant w.r.t. content. This remains an open problem at the source of active
research activities.
In [9], statistical models for components of the video structure are introduced
to classify video sequences into different genres. The analysis of image motion
is widely exploited for the segmentation of videos into meaningful units or for
event recognition. Efficient motion characterization can be derived from the op-
tical flow, as in [8] for human action change detection. In [11], the authors use
very simple local spatio-temporal measurements, i.e., histograms of the spatial
and temporal intensity gradients, to cluster temporal dynamic events. In [10], a
principal component representation of activity parameters (such as translation,
T. Pajdla and J. Matas (Eds.): ECCV 2004, LNCS 3023, pp. 145–157, 2004.
c Springer-Verlag Berlin Heidelberg 2004
