A biologically-inspired vision architecture for resource-constrained intelligent vehicles

Michalke, Thomas; Fritsch, Jannik; Goerick, Christian

doi:10.1016/j.cviu.2009.12.007

Cited by 7 publications

(6 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Top-down attention approaches such as [ 6 , 23 ] differ from our work by their way of acquiring models. Although these authors present integrated systems using object-specific attentional modulation, such modulation is obtained by performing an offline optimization based on heuristically defined positive and negative examples.…”

Section: Discussionmentioning

confidence: 99%

“…Whereas the work of [23] focuses on car detection in road traffic scenarios, the VOCUS model [6] targets mobile robotics applications. Both approaches use an offline optimization procedure to generate feature-based object search templates based on small numbers of image patches.…”

Section: Introductionmentioning

confidence: 99%

“…Another group of attention models focuses on feature-based, object-specific selectivity through learned search models, as well as applicability in real-world scenarios. Whereas the work of [ 23 ] focuses on car detection in road traffic scenarios, the VOCUS model [ 6 ] targets mobile robotics applications. Both approaches use an offline optimization procedure to generate feature-based object search templates based on small numbers of image patches.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Biased Competition in Visual Processing Hierarchies: A Learning Approach Using Multiple Cues

et al. 2011

Self Cite

View full text Add to dashboard Cite

In this contribution, we present a large-scale hierarchical system for object detection fusing bottom-up (signal-driven) processing results with top-down (model or task-driven) attentional modulation. Specifically, we focus on the question of how the autonomous learning of invariant models can be embedded into a performing system and how such models can be used to define object-specific attentional modulation signals. Our system implements bi-directional data flow in a processing hierarchy. The bottom-up data flow proceeds from a preprocessing level to the hypothesis level where object hypotheses created by exhaustive object detection algorithms are represented in a roughly retinotopic way. A competitive selection mechanism is used to determine the most confident hypotheses, which are used on the system level to train multimodal models that link object identity to invariant hypothesis properties. The top-down data flow originates at the system level, where the trained multimodal models are used to obtain space- and feature-based attentional modulation signals, providing biases for the competitive selection process at the hypothesis level. This results in object-specific hypothesis facilitation/suppression in certain image regions which we show to be applicable to different object detection mechanisms. In order to demonstrate the benefits of this approach, we apply the system to the detection of cars in a variety of challenging traffic videos. Evaluating our approach on a publicly available dataset containing approximately 3,500 annotated video images from more than 1 h of driving, we can show strong increases in performance and generalization when compared to object detection in isolation. Furthermore, we compare our results to a late hypothesis rejection approach, showing that early coupling of top-down and bottom-up information is a favorable approach especially when processing resources are constrained.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Biased Competition in Visual Processing Hierarchies: A Learning Approach Using Multiple Cues

et al. 2011

Self Cite

View full text Add to dashboard Cite

show abstract

“…Regarding the latter, RADAR has a narrow field of view and as they are active sensors, they may interfere with other RADAR emitting sources including vehicles with the same technology on board; LASERS are affected by fog and heavy rain; ultrasounds easily fail due to lateral wind; and vision systems can be affected by the changing lighting of outdoor applications and the image processing involves a large computational load which can compromise the real-time performance. Due to these facts, some approaches have focused on a combination of some of the above in order to achieve more reliability [9][10][11].…”

Section: Related Workmentioning

confidence: 99%

Monovision‐based vehicle detection, distance and relative speed measurement in urban traffic

Arenado

Oria

Torre-Ferrero

et al. 2014

IET Intelligent Transport Systems

View full text Add to dashboard Cite

Abstract-This paper presents a monovision-based system for on-road vehicle detection and computation of distance and relative speed in urban traffic. Many works have dealt with monovision vehicle detection, but only a few of them provide the distance to the vehicle which is essential for the control of an Intelligent Transportation System (ITS). The system proposed integrates a single camera reducing the monetary cost of stereovision and RADAR-based technologies. The algorithm is divided in three major stages. For vehicle detection we use a combination of two features: the shadow underneath the vehicle and horizontal edges. We propose a new method for shadow thresholding based on the grayscale histogram assessment of a region of interest on the road. In the second and third stages, the vehicle hypothesis verification and the distance are obtained by means of its number plate whose dimensions and shape are standardized in each country. The analysis of consecutive frames is employed to calculate the relative speed of the vehicle detected. Experimental results showed excellent performance in both vehicle and number plate detections and in the distance measurement, in terms of accuracy and robustness in complex traffic scenarios and under different lighting conditions.

show abstract

“…Throughout the last decade, computational models of attention have attracted an increasing interest in theory (e.g., [4,14,21,25,27]) and applications such as, e.g., humanrobot interaction (e.g., [22]), scene exploration and analysis (e.g., [16]), and driver assistance (e.g., [17]). To this end, several computational models of attention have been developed (e.g., [1, 10-14, 30, 31]).…”

Section: Related Workmentioning

confidence: 99%

Predicting human gaze using quaternion DCT image signature saliency and face detection

Schauerte

Stiefelhagen

2012

2012 IEEE Workshop on the Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

We combine and extend the previous work on DCT-based image signatures and face detection to determine the visual saliency. To this end, we transfer the scalar definition of image signatures to quaternion images and thus introduce a novel saliency method using quaternion type-II DCT image signatures. Furthermore, we use MCT-based face detection to model the important influence of faces on the visual saliency using rotated elliptical Gaussian weight functions and evaluate several integration schemes. In order to demonstrate the performance of the proposed methods, we evaluate our approach on the Bruce-Tsotsos (Toronto) [2] and Cerf (FIFA) [3] benchmark eye-tracking data sets. Additionally, we present evaluation results on the BruceTsotsos data set of the most important spectral saliency approaches. We achieve state-of-the-art results in terms of the well-established area under curve (AUC) measure on the Bruce-Tsotsos data set and come close to the ideal AUC on the Cerf data set -with less than one millisecond to calculate the bottom-up QDCT saliency map.

show abstract

A biologically-inspired vision architecture for resource-constrained intelligent vehicles

Cited by 7 publications

References 42 publications

Biased Competition in Visual Processing Hierarchies: A Learning Approach Using Multiple Cues

Biased Competition in Visual Processing Hierarchies: A Learning Approach Using Multiple Cues

Monovision‐based vehicle detection, distance and relative speed measurement in urban traffic

Predicting human gaze using quaternion DCT image signature saliency and face detection

Contact Info

Product

Resources

About