Model-based recognition of intersections and lane structures

Gengenbach, V.; Nagel, Hans-Hellmut; Heimes, F.; Struck, G.; Kollnig, Henner

doi:10.1109/ivs.1995.528334

Cited by 14 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For a long time intersection understanding has been recognized as a difficult problem [16], [22], [24], [35], [53]. For instance, Luetzeler and Dickmanns [48] extract local image features and match these to a T-shaped intersection model that involves several parameters.…”

Section: Related Workmentioning

confidence: 99%

3D Traffic Scene Understanding From Movable Platforms

Geiger

Lauer

Wojek

et al. 2014

IEEE Trans. Pattern Anal. Mach. Intell.

392

197

View full text Add to dashboard Cite

Abstract-In this paper, we present a novel probabilistic generative model for multi-object traffic scene understanding from movable platforms which reasons jointly about the 3D scene layout as well as the location and orientation of objects in the scene. In particular, the scene topology, geometry and traffic activities are inferred from short video sequences. Inspired by the impressive driving capabilities of humans, our model does not rely on GPS, lidar or map knowledge. Instead, it takes advantage of a diverse set of visual cues in the form of vehicle tracklets, vanishing points, semantic scene labels, scene flow and occupancy grids. For each of these cues we propose likelihood functions that are integrated into a probabilistic generative model. We learn all model parameters from training data using contrastive divergence. Experiments conducted on videos of 113 representative intersections show that our approach successfully infers the correct layout in a variety of very challenging scenarios. To evaluate the importance of each feature cue, experiments using different feature combinations are conducted. Furthermore, we show how by employing context derived from the proposed method we are able to improve over the state-of-the-art in terms of object detection and object orientation estimation in challenging and cluttered urban environments.

show abstract

Section: Related Workmentioning

confidence: 99%

3D Traffic Scene Understanding From Movable Platforms

Geiger

Lauer

Wojek

et al. 2014

IEEE Trans. Pattern Anal. Mach. Intell.

392

197

View full text Add to dashboard Cite

show abstract

“…Several previous approaches (e.g. [3,10,13]) incorporate model information in order to improve the performance of their algorithms. But the term "model" in the field of road detection is not unambiguous, e.g.…”

Section: Figure 3: Dynamic Windows Approachmentioning

confidence: 99%

“…In Gengenbach et al [10] a geometrical model is extracted from apriori digital maps from commercial automatic navigation systems. The model contains information about the lane structure, lane widths and distances between road junctions and intersections.…”

Section: Roadmentioning

confidence: 99%

Towards an approach for knowledge-based road detection

Foedisch¹,

Schlenoff²,

Shneier³

2005

Proceedings of the 2005 ACM Workshop on Research in Knowledge Representation for Autonomous Systems

View full text Add to dashboard Cite

Our previous work on road detection suggests the usage of prior knowledge in order to improve performance. In this paper we will explain our motivation for a novel approach, define requirements and point out issues, particularly concerning the representation of road depending on the use, which need to be addressed. The proposed system will provide symbolic data for high-level processes and guidance for low-level processes. Furthermore, we will outline the recognition approach based on previously discussed requirements and issues. This paper has a visionary character based on our experience with road detection for autonomous road vehicles.

show abstract

“…In contrast to our approach, [16], [17], [13] exploit digital road maps to generate road models which are mapped The task of detecting intersections is aggravated by the diversity in scene appearance, the tilted installation angle of the camera and large occlusions, which mainly caused by other traffic participants. Approaches solely based on lane markings will fail in such situations.…”

Section: Introductionmentioning

confidence: 99%

Object flow: A descriptor for classifying traffic motion

Geiger

Kitt

2010

2010 IEEE Intelligent Vehicles Symposium

View full text Add to dashboard Cite

Abstract-We present and evaluate a novel scene descriptor for classifying urban traffic by object motion. Atomic 3D flow vectors are extracted and compensated for the vehicle's egomotion, using stereo video sequences. Votes cast by each flow vector are accumulated in a bird's eye view histogram grid. Since we are directly using low-level object flow, no prior object detection or tracking is needed. We demonstrate the effectiveness of the proposed descriptor by comparing it to two simpler baselines on the task of classifying more than 100 challenging video sequences into intersection and non-intersection scenarios. Our experiments reveal good classification performance in busy traffic situations, making our method a valuable complement to traditional approaches based on lane markings.

show abstract

Model-based recognition of intersections and lane structures

Cited by 14 publications

References 12 publications

3D Traffic Scene Understanding From Movable Platforms

3D Traffic Scene Understanding From Movable Platforms

Towards an approach for knowledge-based road detection

Object flow: A descriptor for classifying traffic motion

Contact Info

Product

Resources

About