Constrained key-frame based local bundle adjustment is at the core of many recent systems that address the problem of large-scale, georeferenced SLAM based on a monocular camera and on data from inexpensive sensors and/or databases. The majority of these methods, however, impose constraints that result from proprioceptive sensors (e.g. IMUs, GPS, Odometry) while ignoring the possibility of explicitly constraining the structure (e.g. point cloud) resulting from the reconstruction process. Moreover, research on on-line interactions between SLAM and deep learning methods remains scarce, and as a result, few SLAM systems take advantage of deep architectures. We explore both these areas in this work: we use a fast deep neural network to infer semantic and structural information about the environment, and using a Bayesian framework, inject the results into a bundle adjustment process that constrains the 3d point cloud to texture-less 3d building models.

We address the issue of real-time pedestrians detection in a urban environment. This is a challenging task owing to the high variability of appearances and poses that humans can have and to the complexity of backgrounds. We propose a solution made of gradient-based local descriptors combined to form strong classifiers and organized in a cascaded detector. We developed for this an extension of the Histograms of Oriented Gradients (HOGs) and added a new component to the histogram which represents the strength of edges or the amount of information in the histogram support. We also implemented a learning algorithm based on Real Adaboost where two phases – selection first, then refinement of weights – provide more robustness to the detector. We evaluated our system by comparing it to the cascaded detector of Haar features of Viola & Jones [7] and to the SVM of HOGs features of Dalal & Triggs [1]. To ensure an equitable and valid comparison, we used the database proposed in [1]. Our system outperforms them in detection results and in time needs. 1

This article describes a video-surveillance system developed within the ISCAPS project. Thermal imaging provides a robust solution to visibility change (illumination, smoke) and is a relevant technology for discriminating humans in complex scenes. In this article, we demonstrate its efficiency for posture analysis in dense groups of people. The objective is to automatically detect several persons lying down in a very crowded area. The presented method is based on the detection and segmentation of individuals within groups of people using a combination of several weak classifiers. The classification of extracted silhouettes enables to detect abnormal situations. This approach was successfully applied to the detection of terrorist gas attacks on railway platform and experimentally validated in the project. Some of the results are presented here.

A strategy combining advantages of view-based and model-based object recognition approaches has been developed. Textured 3D models are used to produce local appearances of object key-points. Correspondences between 3D points and local appearances are also established during this learning stage. The recognition consists first in matching these local appearances and those extracted from the image. A robust 3D pose estimation is then carried out, discarding spurious correspondences. This algorithm is computationally very efficient. It can analyse images including several objects and can also handle partial object occultations as well as important changes of brightness and contrast.