Internship proposal

Multisensor Analysis for Human Activity Recognition

Keywords

multisensor analysis of human behaviors, temporal scenario recognition, video sequence interpretation.

Internship Context and Goals

The scope of the internship is in agreement with the ongoing research of the ORION group. It concerns automatic interpretation of video sequences pour human behavior recognition.

The ORION team has developed video sequence interpretation platform capable of detecting and tracking moving objects and recognizing certain behaviors that are predefined by experts in a database of scenarii. Especially, the platform has been tested for various applications such as metro and bank surveillance.

In addition to the a priori knowledge about the scenarii to be recognized and the environment observed, at each instant, the recognition process takes as input: (1) a set of individuals being tracked by a vision module and (2) a set of scenarii already recognized. The output of the process is a set of scenarii recognized for the current instant. A scenario is a model of the behavior of one or several individuals evolving in a scene in a given duration. A scenario is composed of a sequence of sub-scenarii meeting some temporal constraints.
For many applications, it is interesting to make use of other information coming from heterogeneous sensors such as cameras, microphones, piezo-electric sensors, etc. This is especially relevant for usual activities in the context of home automation or home nursing of elderly people. It is thus necessary to develop multisensor analysis techniques for human activity monitoring.

Tasks to be accomplished

The first objective of the internship is to propose methods for multisensor analysis of the information about the state of persons being observed. The next objective is to find solutions that are general enough to be integrated into the existing video interpretation platform of the ORION group. The internship consists of both a theoretical study and the implementation of a practical application, such as home nursing of elderly people.
The first step will be a bibliographic study on the subject. The second step will be the development of a new multisensor analysis method of human activities. An implementation of the proposed method should follow and be tested with the VSIP platform, developed by the ORION group.

The internship might lead to a PhD study.

Required Background and Skills

Artificial intelligence, learning techniques, cognitive vision, C++.

Place and duration

6 months within the ORION group of INRIA Sophia-Antipolis.

Supervisors

François BRÉMOND et Monique THONNAT
Projet ORION, INRIA-Sophia Antipolis
2004 route des Lucioles BP 93
06902 Sophia Antipolis Cedex

Contact

Email : Francois.Bremond@sophia.inria.fr
Tel: 04 92 38 76 59

References

[1] Alberto Avanzi, Francois Bremond, Christophe Tornieri and Monique Thonnat, Design and Assessment of an Intelligent Activity Monitoring Platform, in EURASIP Journal on Applied Signal Processing, special issue in "Advances in Intelligent Vision Systems: Methods and Applications", 2005.
[2] B. Georis, F. Brémond, M. Thonnat and B. Macq, Use of an Evaluation and Diagnosis Method to Improve Tracking Performances . The 3rd IASTED International Conference on Visualization, Imaging and Image Proceeding VIIP 03 , September 8-10 2003, Benalmadera, Spain.
[3] Van-Thinh Vu, François Brémond et Monique Thonnat. Automatic Video Interpretation: A Novel Algorithm for Temporal Scenario Recognition. The Eighteenth International Joint Conference on Artificial Intelligence (IJCAI'03), Acapulco, Mexico, August 2003.