Frédéric Champagnat

Learn More
We study dense optical flow estimation using iterative registration of local window, also known as iterative Lucas-Kanade (LK) [B. Lucas et al, 1981]. We show that the usual iterative-warping scheme encounters divergence problems and propose a modified scheme with better behavior. It yields good results with a much lower cost than the exact dense LK(More)
Iteratively Reweighted Least Squares (IRLS) and Residual Steepest descent (RSD) algorithms of robust statistics arise as special cases of half-quadratic schemes . Here, we adopt a statistical framework and we show that both algorithms are instances of the EM algorithm. The augmented dataset respectively involves a scale and a location mixture of Gaussians.(More)
Super-resolution (SR) techniques make use of subpixel shifts between frames in an image sequence to yield higher resolution images. We propose an original observation model devoted to the case of nonisometric inter-frame motion as required, for instance, in the context of airborne imaging sensors. First, we describe how the main observation models used in(More)
—This paper provides a complete characterization of stationary Markov random fields on a finite rectangular (non-toroidal) lattice in the basic case of a second-order neighborhood system. Equivalently, it characterizes stationary Markov fields on 2 whose restrictions to finite rectangular subsets are still Markovian (i.e., even on the boundaries). Until(More)
This paper presents an original answer to the difficult problem of multitarget detection and tracking with aerial images in an urban context. This framework multiplies difficulties for object detection as well as for tracking. Concerning detection, we have to deal with camera motion, strong parallax effects in urban areas, low image resolution, and unknown(More)
This paper deals with dense optical flow estimation from the perspective of the trade-off between quality of the estimated flow and computational cost which is required by real-world applications. We propose a fast and robust local method, denoted by eFOLKI, and describe its implementation on GPU. It leads to very high performance even on large image(More)
Because true Maximum Likelihood (ML) is too expensive, the dominant approach in Bernoulli-Gaussian (BG) myopic deconvolution consists in the joint maximization of a single Generalized Likelihood with respect to the input signal and the hyperparameters. This communication assesses the theoretical properties of a related Maximum Generalized Marginal(More)
We address DSM reconstruction from calibrated limited-angle aerial side-looking image sequences. We use a regularised approach which combines a multi-view pixel-wise similarity criterion and a L1-norm regularisation term. Although it gives quite good results, it has two main drawbacks: occlusions are not dealt with and the reconstruction improvement brought(More)