Appearance-based Brake-Lights recognition using deep learning and vehicle detection

Wang, Jiangang; Zhou, Lin; Pan, Yu; Lee, Serin; Song, Zhiwei; Han, Boon Siew; Saputra, Vincensius Billy

doi:10.1109/ivs.2016.7535481

Cited by 68 publications

(39 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhong et al [12] train a fully convolutional network (FCN) [23] model to identify the light regions and the features extracted within the regions are classified by a linear SVM. Wang et al [4] train a CNN model from vehicle rear appearances to tell the state of the brake signals image by image. In order to take temporal dependencies into account, Hsu et al [5] propose a CNN-LSTM structure to learn eight states of taillights, where the networks for brake and turn signals are trained separately.…”

Section: A Vehicle Taillight Recognitionmentioning

confidence: 99%

“…{kuan.lee, takaaki.tagawa, marcus.pan, adrien.gaidon, bertrand.douillard}@tri.global fusion architecture is proposed to fuse spatial and temporal dependencies together [18] [19]. Such deep learning methods have achieved promising results, inspiring the researchers to apply the ideas and the techniques to vehicle taillight recognition [4] [5].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Attention-based Recurrent Convolutional Network for Vehicle Taillight Recognition

Lee¹,

Tagawa²,

Pan³

et al. 2019

2019 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Vehicle taillight recognition is an important application for automated driving, especially for intent prediction of ado vehicles and trajectory planning of the ego vehicle. In this work, we propose an end-to-end deep learning framework to recognize taillights, i.e. rear turn and brake signals, from a sequence of images. The proposed method starts with a Convolutional Neural Network (CNN) to extract spatial features, and then applies a Long Short-Term Memory network (LSTM) to learn temporal dependencies. Furthermore, we integrate attention models in both spatial and temporal domains, where the attention models learn to selectively focus on both spatial and temporal features. Our method is able to outperform the state of the art in terms of accuracy on the UC Merced Vehicle Rear Signal Dataset, demonstrating the effectiveness of attention models for vehicle taillight recognition.

show abstract

Section: A Vehicle Taillight Recognitionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An Attention-based Recurrent Convolutional Network for Vehicle Taillight Recognition

Lee¹,

Tagawa²,

Pan³

et al. 2019

2019 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

show abstract

“…This method needs to further improve the matching effect of the taillights and solve the problem of poor detection of the red vehicles. Wang et al used a vehicle's rear appearance image to learn "brake light mode" through a multi-layer perceptron neural network in a large database and trained a depth classifier to judge whether the taillight was in the normal or braking state [14]. Wang et al proposed a method to optimize the faster RCNN for vehicle detection by improving multi shapes receptive field generation, anchor generation optimization and ROI assignment to improve detection speed and accuracy [15].…”

Section: Related Workmentioning

confidence: 99%

Performance Evaluation of Region-Based Convolutional Neural Networks Toward Improved Vehicle Taillight Detection

Wang

Huo

et al. 2019

Applied Sciences

View full text Add to dashboard Cite

Increasingly serious traffic jams and traffic accidents pose threats to the social economy and human life. The lamp semantics of driving is a major way to transmit the driving behavior information between vehicles. The detection and recognition of the vehicle taillights can acquire and understand the taillight semantics, which is of great significance for realizing multi-vehicle behavior interaction and assists driving. It is a challenge to detect taillights and identify the taillight semantics on real traffic road during the day. The main research content of this paper is mainly to establish a neural network to detect vehicles and to complete recognition of the taillights of the preceding vehicle based on image processing. First, the outlines of the preceding vehicles are detected and extracted by using convolutional neural networks. Then, the taillight area in the Hue-Saturation-Value (HSV) color space are extracted and the taillight pairs are detected by correlations of histograms, color and positions. Then the taillight states are identified based on the histogram feature parameters of the taillight image. The detected taillight state of the preceding vehicle is prompted to the driver to reduce traffic accidents caused by the untimely judgement of the driving intention of the preceding vehicle. The experimental results show that this method can accurately identify taillight status during the daytime and can effectively reduce the occurrence of confused judgement caused by light interference.

show abstract

Section: Analysis Of Related Workmentioning

confidence: 99%

“…Deep learning models have the following characteristics: they only consider characteristics such as color or depth, which implies a limitation to obtaining a higher level of abstraction of the representative characteristics of the obstacles [23,18,21]. Tracking methods based on online-classifiers learning suffer the problem of error accumulation during the self-learning process [21,16]. Some algorithms based on the convolutional neural networks approach only capture translational invariances and not the rotational invariance or out-of plane rotation, which makes them susceptible to error when classifying and identifying obstacles [6,23,16].…”

Section: Analysis Of Related Workmentioning

confidence: 99%

Obstacle Detection and Trajectory Estimation in Vehicular Displacements based on Computational Vision

Reyes-Cocoletzi¹,

Olmos-Pineda²,

Olvera-López³

2019

RCS

View full text Add to dashboard Cite

Obstacle detection and trajectory estimation in vehicular environments is an open problem in autonomous vehicles development. The automotive industry has made significant progress in research and development of tools; however, there are still challenges to overcome and opportunity areas to be exploited in order to achieve full autonomy in vehicles. This paper presents an analysis of different methods proposed for obstacle detection and trajectory estimation, leading into a proposal of solution for solving the problem of trajectory estimation based on computer vision techniques. This proposal covers the context of traffic environments in Latin America, where basic signage (such as the dividing lines of the road) is absent or not easily visible, among other typical characteristics of countries (like Mexico and others) where the infrastructure for maintaining safe road conditions (speedbumps, potholes) is limited.

show abstract

Appearance-based Brake-Lights recognition using deep learning and vehicle detection

Cited by 68 publications

References 14 publications

An Attention-based Recurrent Convolutional Network for Vehicle Taillight Recognition

An Attention-based Recurrent Convolutional Network for Vehicle Taillight Recognition

Performance Evaluation of Region-Based Convolutional Neural Networks Toward Improved Vehicle Taillight Detection

Obstacle Detection and Trajectory Estimation in Vehicular Displacements based on Computational Vision

Contact Info

Product

Resources

About