Research

My main research interests are directed along developing statistical models and machine learning algorithms for:

human tracking and activity analysis
image and scene representation modeling
sequence analysis and multimedia content analysis.

Events

NEW: I am co-organising a workshop on Interactive Human Behavior Analysis in Open or Public Spaces which will be held the 16th of November in Amsterdam, as part of the International joint Conference on Ambient Intelligence (AmI).

NEW: One postdoctoral position in computer vision, audio processing, and machine learning for multimodal interaction in robots. See the corresponding page.

Recent publications

Bridging the Past, Present and Future: Modeling Scene Activities From Event Relationships and Global Rules
Jagan Varadarajan, Rémi Emonet and Jean-Marc Odobez,
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Providence, June 2012. Supplementary material.

We are not contortionists: coupled adaptive learning for head and body orientation estimation in surveillance video.
Chen Chen and Jean-Marc Odobez,
Oral presentation at IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Providence, June 2012.

Gaze estimation from multimodal Kinect data.
Kenneth Funes and Jean-Marc Odobez,
Oral presentation at CVPR Workshop on Face and Gesture and Kinect demonstration competition, Providence, June 2012.

Fast Human Detection from Joint Appearance and Foreground Feature Subset Covariances
J. Yao and J.-M. Odobez,
Computer Vision and Image Understanding, Vol. 115(10), pp. 1414-1426, October 2011.
Research report with more details about the method and results

Joint Adaptive Colour Modelling and Skin, Hair and Clothing Segmentation Using Coherent Probabilistic Index Maps
Carl Scheffler and Jean-Marc Odobez
British Machine Vision Conference (BMVC), Aberdeen, August 2011.
Pdf of Main article + supplementary material ------ Webpage of corresponding code and data

Combined Estimation of Location and Body Pose in Surveillance Video
Cheng Chen, Alexandre Heili, and Jean-Marc Odobez,
IEEE Int. Conf. on Advanced Video and Signal-Based Surveillance (AVSS), Klagenfurt, Austria, Aug. 2011.

Multi-camera Open Space Human Activity Discovery for Anomaly Detection
Rémi Emonet, Jagan Varadarajan and Jean-Marc Odobez,
IEEE Int. Conf. on Advanced Video and Signal-Based Surveillance (AVSS), Klagenfurt, Austria, Aug. 2011.

Extracting and Locating Temporal Motifs in Video Scenes Using a Hierarchical Non Parametric Bayesian Model
Rémi Emonet, Jagan Varadarajan and Jean-Marc Odobez,
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Colorado Springs, June 2011.

Exploiting Long-Term Observations for Track Creation and Deletion in Online Multi-Face Tracking
S. Duffner and J.-M. Odobez,
IEEE Conference on Automatic Face and Gesture Recognition, Santa Barbara, March 2011.

Probabilistic Latent Sequential Motifs: Discovering temporal activity patterns in video scenes
Jagan Varadarajan, Rémi Emonet and Jean-Marc Odobez,
in Proc. British Machine and Vision Conference (BMVC), Aberystwyth, Sept. 2010.

Multi-Person Visual Focus of Attention from Head Pose and Meeting Contextual Cues
S. Ba and J.-M. Odobez,
IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI), Vol. 33. No. 1. pp. 101-116, January 2011.

View-Based Appearance Model Online Learning for 3D Deformable Face Tracking
Stéphanie Lefèvre and Jean-Marc Odobez,
in Proc. Int. Conf. on Computer Vision Theory and Applications (VISAPP), Angers, May 2010.

Multi-Camera Multi-Person 3D Space Tracking with MCMC in Surveillance Scenarios
Jian Yao and Jean-Marc Odobez,
in European Conference on Computer Vision, workshop on Multi Camera and Multi-modal Sensor Fusion Algorithms and Applications (ECCV-M2SFA2), Marseille, Oct. 2008.