Abstract — Pedestrian detection is a rapidly evolving area in the intelligent vehicles domain. Stereo vision is an attractive sensor for this purpose. But unlike for monocular vision, there are no realistic, large scale benchmarks available for stereobased pedestrian detection, to provide a common point of reference for evaluation. This paper introduces the Daimler Stereo-Vision Pedestrian Detection benchmark, which consists of several thousands of pedestrians in the training set, and a 27-min test drive through urban environment and associated vehicle data. The data, including ground truth, is made publicly available for non-commercial purposes. The paper furthermore quantifies the benefit of stereo vision for ROI generation and localization; at equal detection rates, false positives are reduced by a factor of 4-5 with stereo over mono, using the same HOG/linSVM classification component. I.

This paper presents a novel approach to single-frame pedestrian classification and orientation estimation. Unlike previous work which addressed classification and orientation separately with different models, our method involves a probabilistic framework to approach both in a unified fashion. We address both problems in terms of a set of view-related models which couple discriminative expert classifiers with sample-dependent priors, facilitating easy integration of other cues (e.g. motion, shape) in a Bayesian fashion. This mixture-of-experts formulation approximates the probability density of pedestrian orientation and scalesup to the use of multiple cameras. Experiments on large real-world data show a significant performance improvement in both pedestrian classification and orientation estimation of up to 50%, compared to stateof-the-art, using identical data and evaluation techniques. 1.

This paper introduces active pedestrian protection system (APPS) project, which was fulfilled during last three years (Nov. 2006 – Oct. 2009) and supported by ministry of knowledge economy (MKE) of Korean government, and then summarizes its achievements. To reduce the number and severity of vehicle-pedestrian accident by active safety vehicle (ASV), false alarm rate should be minimized. For this purpose, we developed critical area-centered pedestrian detection system which recognized locations of parked vehicles along the road-side then investigated only areas behind the parked vehicles. After evaluating four different approaches, sensor fusion of near infrared (NIR) vision and scanning laser radar was selected as the optimal. Although feasibility of braking and suspension for pedestrian protection was evaluated, only braking was proved to be useful.

Abstract—This paper presents a novel pedestrian detection system for intelligent vehicles. We propose the use of dense stereo for both the generation of regions of interest and pedestrian classification. Dense stereo allows the dynamic estimation of camera parameters and the road profile, which, in turn, provides strong scene constraints on possible pedestrian locations. For classification, we extract spatial features (gradient orientation histograms) directly from dense depth and intensity images. Both modalities are represented in terms of individual feature spaces, in which discriminative classifiers (linear support vector machines) are learned. We refrain from the construction of a joint feature space but instead employ a fusion of depth and intensity on the classifier level. Our experiments involve challenging image data captured in complex urban environments (i.e., undulating roads and speed bumps). Our results show a performance improvement by up to a factor of 7.5 at the classification level and up to a factor of 5 at the tracking level (reduction in false alarms at constant detection rates) over a system with static scene constraints and intensity-only classification. Index Terms—Active safety, computer vision, intelligent vehicles, pedestrian detection.

Abstract — This paper introduces a part-based two-stage pedestrian detector. The system finds pedestrian candidates with an AdaBoost cascade on Haar-like features. It then verifies each candidate using a part-based HOG-SVM doing first a regression and then a classification based on the estimated function output from the regression. It uses the Histogram of Oriented Gradients (HOG) computed on both the full, upper and lower body of the candidates, and uses these in the final verification. The system has been trained and tested on the INRIA dataset and performs better than similar previous work, which uses full-body verification. I.

Abstract. Road scene segmentation is important in computer vision for different applications such as autonomous driving and pedestrian detection. Recovering the 3D structure of road scenes provides relevant contextual information to improve their understanding. In this paper, we use a convolutional neural network based algorithm to learn features from noisy labels to recover the 3D scene layout of a road image. The novelty of the algorithm relies on generating training labels by applying an algorithm trained on a general image dataset to classify on–board images. Further, we propose a novel texture descriptor based on a learned color plane fusion to obtain maximal uniformity in road areas. Finally, acquired (off–line) and current (on–line) information are combined to detect road areas in single images. From quantitative and qualitative experiments, conducted on publicly available datasets, it is concluded that convolutional neural networks are suitable for learning 3D scene layout from noisy labels and provides a relative improvement of 7 % compared to the baseline. Furthermore, combining color planes provides a statistical description of road areas that exhibits maximal uniformity and provides a relative improvement of 8 % compared to the baseline. Finally, the improvement is even bigger when acquired and current information from a single image are combined. 1

Abstract not found