Elise Arnaud

Motion Compression using Principal Geodesics Analysis

Due to the growing need for large quantities of human animation data in the entertainment industry, it has become a necessity to compress motion capture sequences in order to ease their storage and transmission. We present a novel, lossy compression method for human motion data that exploits both temporal and spatial coherence. Given one motion, we first approximate the poses manifold using Principal Geodesics Analysis (PGA) in the configuration space of the skeleton. We then search this approximate manifold for poses matching end‐effectors constraints using an iterative minimization algorithm that allows for real‐time, data‐driven inverse kinematics. The compression is achieved by only storing the approximate manifold parametrization along with the end‐effectors and root joint trajectories, also compressed, in the output data. We recover poses using the IK algorithm given the end‐effectors trajectories. Our experimental results show that considerable compression rates can be obtained using our method, with few reconstruction and perceptual errors.

Conditional filters for image sequence-based tracking - application to point tracking

A new conditional formulation of classical filtering methods is proposed. This formulation is dedicated to image sequence-based tracking. These conditional filters allow solving systems whose measurements and state equation are estimated from the image data. In particular, the model that is considered for point tracking combines a state equation relying on the optical flow constraint and measurements provided by a matching technique. Based on this, two point trackers are derived. The first one is a linear tracker well suited to image sequences exhibiting global-dominant motion. This filter is determined through the use of a new estimator, called the conditional linear minimum variance estimator. The second one is a nonlinear tracker, implemented from a conditional particle filter. It allows tracking of points whose motion may be only locally described. These conditional trackers significantly improve results in some general situations. In particular, they allow for dealing with noisy sequences, abrupt changes of trajectories, occlusions, and cluttered background.

A fluid motion estimator for schlieren image velocimetry

In this paper, we address the problem of estimating the motion of fluid flows that are visualized through a Schlieren system. Such a system is well known in fluid mechanics as it enables the visualization of unseeded flows. As the resulting images exhibit very low photometric contrasts, classical motion estimation methods based on the brightness consistency assumption (correlation-based approaches, optical flow methods) are completely inefficient. This work aims at proposing a sound energy based estimator dedicated to these particular images. The energy function to be minimized is composed of (a) a novel data term describing the fact that the observed luminance is linked to the gradient of the fluid density and (b) a specific div curl regularization term. The relevance of our estimator is demonstrated on real-world sequences.

An efficient Rao-Blackwellized particle filter for object tracking

In this paper we present a technique for the tracking of textured almost planar object. The target is modeled as a noisy planar cloud of points. The tracking is led with an appropriate non linear stochastic filter. The particular system that we devised is conditionally Gaussian and can be efficiently implemented through variance reduction principle known as Rao-Blackwellisation. Our model allows also to melt a correlation measurements with dynamic model estimated from the images. Such a cooperation within a stochastic filtering framework allows the tracker to be robust to occlusions and targets unpredictable changes of speed and direction. We demonstrate the efficiency of the tracker on different types of real world sequences.

A robust and automatic face tracker dedicated to broadcast videos

Because of their lack of rules, general broadcast videos are more difficult to analyze than news or sport videos. To retrieve human interventions in this context, a robust face tracker is needed. The approach we investigate for face tracking combines three main modules that are a face detector, a region-based tracker and an eye tracker. The region-based tracker relies on a robust parametric motion estimation technique. The eye tracker is based on a Kalman filter. The analysis of the coherence of the trackers output provides an efficient way to detect profile positions and tracking errors. We have thus defined an entirely automatic tracker, able to manage several appearing/disappearing faces, without any a priori knowledge on the image sequence. Experimental results on broadcast videos demonstrate its efficiency to deal with large and rapid motions, occlusions and faces in profile position.

Cooperative disparity and object boundary estimation

In this paper we carry out cooperatively both disparity and object boundary estimation by setting the two tasks in a unified Markovian framework. We introduce a new joint probabilistic model that allows to estimate disparities through a Markov random field model. Boundary estimation then cooperates with disparity estimation to gradually and jointly improve accuracy. The feedback from boundary estimation to disparity estimation is made through the use of an auxiliary field referred to as a displacement field. This field suggests the corrections that need to be applied at disparity discontinuities in order that they align with object boundaries. The joint model reduces to a Markov random field model when considering disparities while it reduces to a Markov chain when focusing on the displacement field. The performance of our approach is illustrated on real stereo images sets, demonstrating the power of this cooperative framework.

Trains of keypoints for 3D object recognition

This paper presents a 3D object recognition method that exploits the spatio-temporal coherence of image sequences to capture the object most relevant features. We start from an image sequence that describes the objects visual appearance from different view points. We extract local features (SIFT) and track them over the sequence. The tracked interest points form trains of features that are used to build a vocabulary for the object. Training images are represented with respect to that vocabulary and an SVM classifier is trained to recognize the object. We present very promising results on a dataset of 11 objects. Tests are performed under varying illumination, scale, and scene clutter

Disparity and normal estimation through alternating maximization

In this paper, we propose an algorithm that recovers binocular disparities in accordance with the surface properties of the scene under consideration. To do so, we estimate the disparity as well as the normals in the disparity space, by setting the two tasks in a unified framework. A novel joint probabilistic model is defined through two random fields to favor both intra field (within neighboring disparities and neighboring normals) and inter field (between disparities and normals) consistency. Geometric contextual information is introduced in the models for both normals and disparities, which is optimized using an appropriate alternating maximization procedure. We illustrate the performance of our approach on synthetic and real data.