Christian Pere

Improvement of the real-time gesture analysis by a new mother wavelet and the application for the navigation inside a scale-one 3D system

This paper proposes a navigation technique for traveling inside a real-scale 3D model based on human gesture analysis. In the first step, a simple threshold is used as a criterion to analyze gestures. In the next step, a complex criterion will be imposed to the analysis to improve the navigation technique. Walking is a periodic signal and moving feet up and down is a part which is repeated. A mother wavelet is allocated to the selected pattern. Then the position of the pattern is recognized by applying Multi Resolution Analysis (MRAs). The movement command is generated and sent to the graphic render (as a movement command). Analytical results show a very high precision performance in the presence of noise, scale variation and superposition of other signals. Practical experiments also verify the same promising results.

Ray-on, an On-Site Photometric Augmented Reality Device

As buildings from ancient times wither away or are transformed or destroyed, it becomes increasingly difficult to figure what it looked like back then. Virtual reality is a great tool to handle this role, although it often implies a three-dimensional model of a building disconnected from the remains. In this article, we present an on-site device dedicated to architectural heritage, which uses realistic photometric rendering associated with the reproduction of the user’s point of view in the virtual scene to immerse him in the uchronic virtual world while maintaining the link to the real site. This design is already in use at the old church of Cluny (France), where is it was welcomed by tourists and guides.

New motherwavelet for pattern detection in IR image

The paper presents a new mother wavelet adapted from a specific pattern. Wavelet multi-resolution analysis uses this wavelet to detect the position of the pattern in an Infra-Red (IR) signal under scale variation and the presence of noise. IR signal is extracted from IR image sequence recorded by an IR camera, Time of Flight (TOF) sensor configuration. The maximum correlation between the pattern and the signal of interest will be used as a criterion to define the mother wavelet. The proposed mother wavelet were tested and verified under the scale variation and the presence of noise. The experimental tests and performance analysis show promising results for both scale variation and noisy signal. 90% accuracy for the proposed wavelet under intensive noisy condition (50% of the signal amplitude) is guaranteed and high precision is expected under real condition.

Sensor fusion for interactive real-scale modeling and simulation systems

This paper proposes an accurate sensor fusion scheme for navigation inside a real-scale 3D model by combining audio and video signals. Audio signal of a microphone-array is merged by Minimum Variance Distortion-less Response (MVDR) algorithm and processed instantaneously via Hidden Markov Model (HMM) to generate translation commands by word-to-action module of speech processing system. Then, the output of optical head tracker (four IR cameras) is analyzed by non-linear/non-Gaussian Bayesian algorithm to provide information about the orientation of the users head. The orientation is used to redirect the user toward a new direction by applying quaternion rotation. The output of these two sensors (video and audio) is combined under the sensor fusion scheme to perform continuous travelling inside the model. The maximum precision for the traveling task is achieved under sensor fusion scheme. Practical experiment shows promising results for the implementation.