Abstract: Image based human detection is of paramount interest due to its potential applications in fields such as advanced driving assistance, surveillance and media analysis. However, even detecting non-occluded standing humans remains a challenge of intensive research. The most promising human detectors rely on classifiers developed in the discriminative paradigm, i.e., trained with labelled samples. However, labeling is a manual intensive step, especially in cases like human detection where it is necessary to provide at least bounding boxes framing the humans for training. To overcome such problem, some authors have proposed the use of a virtual world where the labels of the different objects are obtained automatically. This means that the human models (classifiers) are learnt using the appearance of rendered images, i.e., using realistic computer graphics. Later, these models are used for human detection in images of the real world. The results of this technique are surprisingly good. However, these are not always as good as the classical approach of training and testing with data coming from the same camera, or similar ones. Accordingly, in this paper we address the challenge of using a virtual world for gathering (while playing a videogame) a large amount of automatically labelled samples (virtual humans and background) and then training a classifier that performs equal, in real-world images, than the one obtained by equally training from manually labelled real-world samples. For doing that, we cast the problem as one of domain adaptation. In doing so, we assume that a small amount of manually labelled samples from real-world images is required. To collect these labelled samples we propose a non-standard active learning technique. Therefore, ultimately our human model is learnt by the combination of virtual and real world labelled samples (Fig. 1), which has not been done before. We present quantitative results showing that this approach is valid.

Abstract: In this paper we improve the histogram of oriented gradients (HOG), a core descriptor of state-of-the-art object detection, by the use of higher-level information coming from image segmentation. The idea is to re-weight the descriptor while computing it without increasing its size. The benefits of the proposal are two-fold: (i) to improve the performance of the detector by enriching the descriptor information and (ii) take advantage of the information of image segmentation, which in fact is likely to be used in other stages of the detection system such as candidate generation or refinement.
We test our technique in the INRIA person dataset, which was originally developed to test HOG, embedding it in a human detection system. The well-known segmentation method, mean-shift (from smaller to larger super-pixels), and different methods to re-weight the original descriptor (constant, region-luminance, color or texture-dependent) has been evaluated. We achieve performance improvements of 4:47% in detection rate through the use of differences of color between contour pixel neighborhoods as re-weighting function.

Abstract: This paper presents a comparative study between two road approximation techniques—planar surfaces—from stereo vision data. The first approach is carried out in the v-disparity space and is based on a voting scheme, the Hough transform. The second one consists in computing the best fitting plane for the whole 3D road data points, directly in the Euclidean space, by using least squares fitting. The comparative study is initially performed over a set of different synthetic surfaces
(e.g., plane, quadratic surface, cubic surface) digitized by a virtual stereo head; then real data obtained with a commercial stereo head are used. The comparative study is intended to be used as a criterion for fining the best technique according to the road geometry. Additionally, it highlights common problems driven from a wrong assumption about the scene’s prior knowledge.

Abstract: This paper presents a robust technique for a real time estimation of both camera’s position and orientation—referred as pose. A commercial stereo vision system is used. Unlike previous approaches, it can be used either for urban or highway scenarios. The proposed technique consists of two stages. Initially, a compact 2D representation of the original 3D data points is computed. Then, a RANSAC based least squares approach is used for fitting a plane to the road. At the same time,
relative camera’s position and orientation are computed. The proposed technique is intended to be used on a driving assistance scheme for applications such as obstacle or pedestrian detection. Experimental results on urban environments with different road geometries are presented.