Abstract

International audience; Vision systems provide a large functional spectrum for perception applications and, in recent years, they have demonstrated to be essential in the development of Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicles. In this context, this paper presents an on-road objects detection ap-proach improved by our previous work in defining the traffic area and new strategy in obstacle extraction from U-disparity.Then, a modified particle filtering is proposed for multiple object tracking . The perception strategy of the proposed vision-only detection system is structured as follows : First, a method based on illuminant invariant image is employed at an early stage for free road space detection. A convex hull is then constructed to generate a region of interest (ROI) which includes the main traffic road area. Based on this ROI, an U-disparity map is built to characterize on-road obstacles. In this approach, connected regions extraction is applied for obstacles detection instead of standard Hough Transform. Finally, a modified particle filter framework is employed for multiple targets tracking based on the former detection results. Besides, multiple cues, such as obstacle's size verification and combination of redundant detections, are em-bedded in the system to improve its accuracy. Our experimental findings demonstrates that the system is effective and reliable when applied on different traffic video sequences from a public database.

Abstract

International audience; Integrity of the information provided by a perception system is crucial for advanced driver assistance systems intended for safety applications, like obstacle avoidance systems. A method to ensure integrity is to use different kinds of perception sources. Lidars are key sensors for multiple objects detection and tracking. Stereo vision systems (SVS) can be used to improve the tracking but, in this paper, we use also SVS to confirm the real existence of potential obstacles thanks to 3D dense reconstruction in focused regions of interest. Synchronization issues between the different sensors are addressed using predictive filtering. The proposed approach is evaluated in real conditions thanks to five use cases relevant to urban situations. Results show that this visual confirmation strategy is efficient.

Abstract

This paper introduces an innovative system easing prototyping and validation of Advanced Driver Assistance Systems (ADAS) for automotive applications. This system, based on a chassis-dynamometer and a multi-sensor simulation software, act as a virtual reality platform for intelligent vehicle. This system allows to test and validate complex ADAS systems under various simulated environmental conditions and scenario without the cost, complexity and hazard associated with test-drives. It can also be used early in the design process to test and validate the integration of multiple ADAS in a vehicle. This system is being used for the Virtual Testing applications in the DESERVE 1 project.

Abstract

International audience; Reliable obstacle detection and localization is a key issue for driver assistance systems, particularly in urban environments. In this study a multi-modal perception approach is investigated, the objective being to enhance vehicle localization and dynamic object tracking in a world-centric map. A 3D ego-localization is achieved by merging information from a stereo vision system and data obtained from vehicle sensors. Mobile objects are detected using a multi-layer lidar that is also used to identify a constrained search space within the multiple target tracking process. Object localization and tracking is then performed in the fixed frame, which facilitates analysis and understanding of the scene. Experimental results using real world data are performed to evaluate the performance of the multi-modal system, and these are presented to show the effectiveness of the approach.

Abstract

Diagnosis of complex systems refers to the problem of identifying a breakdown or a failure based on an inspection, a control or a test. Monitoring such industrial complex systems is essential to schedule relevant maintenance actions. We consider an automotive subsystem to monitor: the brake system, because of its impact on the vehicles availability. Through a European project [1], data are acquired via in-vehicle communication protocols and additional sensors. This work aims at developing remote diagnostic and maintenance support tools driven by these data. Our approach combines an analytic model and detection techniques in order to monitor the brake system. We provide experimental results on vehicle data using two multivariate detection methods.

Abstract

In future Advanced Driver Assistance Systems (ADAS), smart monitoring of the vehicle environment is a key issue. Fisheye cameras have become popular as they provide a panoramic view with a few low-cost sensors. However, current ADAS systems have limited use as most of the underlying image processing has been designed for perspective views only. In this article we illustrate how the theoretical work done in omnidirectional vision over the past ten years can help to tackle this issue. To do so, we have evaluated a simple algorithm for road line detection based on the unified sphere model in real conditions. We firstly highlight the interest of using fisheye cameras in a vehicle, then we outline our method, we present our experimental results on the detection of lines on a set of 180 images, and finally, we show how the 3D position of the lines can be recovered by triangulation.

Abstract

International audience; Understanding driver behaviors is an important need for the Advanced Driver Assistance Systems. In particular, the pedestrian detection systems become extremely distracting and annoying when they inform the driver with unnecessary warning messages. In this paper, we propose to study the driver behaviors whenever a pedestrian appears in front of the vehicle. A method based on the driving actions and the Hidden Markov Model (HMM) algorithm is developed to classify the driver awareness of pedestrian and the driver unawareness of pedestrian. The method is successfully validated using the collected data from the experiments that are conducted on a driving simulator. Furthermore, two simple methods based on the static parameters such as the Time-To-Collision and the Required Deceleration Parameter are also applied to our problem and are compared to the proposed method. The result shows a significant improvement of the HMM-based method compared to the simple ones.

Abstract

International audience; In this work, we investigate the use of OpenStreetMap data for semantic labeling of Earth Observation images. Deep neural networks have been used in the past for remote sensing data classification from various sensors, including multispectral, hyperspectral, SAR and LiDAR data. While OpenStreetMap has already been used as ground truth data for training such networks, this abundant data source remains rarely exploited as an input information layer. In this paper, we study different use cases and deep network architectures to leverage OpenStreetMap data for semantic labeling of aerial and satellite images. Especially , we look into fusion based architectures and coarse-to-fine segmentation to include the OpenStreetMap layer into multispectral-based deep fully convolutional networks. We illustrate how these methods can be successfully used on two public datasets: ISPRS Potsdam and DFC2017. We show that OpenStreetMap data can efficiently be integrated into the vision-based deep learning models and that it significantly improves both the accuracy performance and the convergence speed of the networks.

Abstract

International audience; This paper presents a set of algorithms dedicated to the 3D modeling of historical buildings from a collection of old ar-chitecture plans, including floor plans, elevations and cut-offs. Image processing algorithms help to detect and local-ize main structures of the building from the floor plans (thick and thin walls, openings). The extrusion of the walls allow us to build a first 3D model. We compute height informations and add textures to the model by analyzing the elevation im-ages from the same collection of documents. We applied this pipeline to XVIII th century plans of the Château de Versailles, and show results for two different parts of the Château.

Abstract

International audience; The estimation of disparity maps from stereo pairs has many applications in robotics and autonomous driving. Stereo matching has first been solved using model-based approaches, with real-time considerations for some, but to-day's most recent works rely on deep convolutional neural networks and mainly focus on accuracy at the expense of computing time. In this paper, we present a new method for disparity maps estimation getting the best of both worlds: the accuracy of data-based methods and the speed of fast model-based ones. The proposed approach fuses prior disparity maps to estimate a refined version. The core of this fusion pipeline is a convolutional neural network that leverages dilated convolutions for fast context aggregation without spatial resolution loss. The resulting architecture is both very effective for the task of refining and fusing prior disparity maps and very light, allowing our fusion pipeline to produce disparity maps at rates up to 125 Hz. We obtain state-of-the-art results in terms of speed and accuracy on the KITTI benchmarks. Code and pre-trained models are available on our github: https://github.com/ ferreram/FD-Fusion.