System approach for multi-purpose representations of traffic scene elements

Abstract

Abstract-A major step towards intelligent vehicles lies in the acquisition of an environmental representation of sufficient generality to serve as the basis for a multitude of different assistance-relevant tasks. This acquisition process must reliably cope with the variety of environmental changes inherent to traffic environments. As a step towards this goal, we present our most recent integrated system performing object detection in challenging environments (e.g., inner-city or heavy rain). The system integrates unspecific and vehicle-specific methods for the detection of traffic scene elements, thus creating multiple object hypotheses. Each detection method is modulated by optimized models of typical scene context features which are used to enhance and suppress hypotheses. A multi-object tracking and fusion process is applied to make the produced hypotheses spatially and temporally coherent. In extensive evaluations we show that the presented system successfully analyzes scene elements under diverse conditions, including challenging weather and changing scenarios. We demonstrate that the used generic hypothesis representations allow successful application to a variety of tasks including object detection, movement estimation, and risk assessment by time-to-contact evaluation.

Robust visual pedestrian detection by tight coupling to tracking

“…Most approaches include tracking, usually in various flavors of Kalman or particle filtering [13], [14], [15], [16], [17], [18], [19]. Both tracking methods perform a kind of "late fusion", combining detection results with a motion model, the latter having no infleucne at all on detection process.…”

The contribution of context information: A case study of object recognition in an intelligent car

“…The coupling of object detection and contextual information mediated by low-level modulation is demonstrated in [27] where "gist", a low-dimensional vi- Figure 1: Two different ways of incorporating context models into SamSys [5,26]. Left: data flow schematics for both cases, either attentional modulation of low-level processing (case a) or purely high-level hypothesis selection (case b).…”

“…Prominent examples for this development are, e.g., humanoid robots [2,3] and intelligent vehicles [4,5]. Numerous ways exist for coping with this additional complexity, such as formal hierarchical design languages [6], component-based graphical development systems [7] or machine learning methods [8,9].…”

Autonomous Generation of Internal Representations for Associative Learning

“…A car is equipped with two front cameras and a laser in order to acquire data which is used by several processing layers [3]. The output of these layers is converted into population-codes (PCs) following the method described in Sec 2.1.…”

“…This study is conducted based on an instance of a large-scale object detection system in road traffic environments [3] which integrates multimodal information (laser, video) as well as a wide variety of vision-based algorithms such as stereo, tracking, classification, and free-area detection. The motivation for this study arose when trying to obtain multimodal object models (in this case: car models) for excluding obviously incorrect object detections.…”