Vehicle Detection With Automotive Radar Using Deep Learning on Range-Azimuth-Doppler Tensors

Major, Bence; Fontijne, Daniel; Ansari, Amin; Sukhavasi, Ravi Teja; Radhika, G; Hamilton, Michael; Lee, Sean; Grzechnik, Slawomir K.; Subramanian, S.

doi:10.1109/iccvw.2019.00121

Cited by 154 publications

(87 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Wöhler et al [13, 14] used Long Short‐Term Memory (LSTM) neural networks to classify road actors in the automotive scenario in which the motion‐compensated Doppler velocity was a key feature. Other broadly similar works include Rohling et al [15], who used a 24 GHz radar to classify pedestrians by analysing the Doppler spectrum and range profile, Major et al [16] who classified and detected vehicles in a highway scenario using a range‐azimuth‐Doppler spectrum based on 3D convolutions and LSTM networks, and Bartsch et al [17] who classified pedestrians using the area and shape of the object and Doppler spectrum features. Bartsch et al analysed the probability of each feature and used a simple decision model, achieving 95% accurate classification rates for optimal scenarios, but this dropped to 29.4% when the pedestrian was in close proximity to cars due to low resolution from the radar sensors.…”

Section: Related Workmentioning

confidence: 99%

300 GHz radar object recognition based on deep neural networks and transfer learning

Sheeny

Wallace

Wang

2020

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

300 GHz radar object recognition based on deep neural networks and transfer learning

Sheeny

Wallace

Wang

2020

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

show abstract

“…Finally, a comprehensive analysis of applying deep learning to radar signals is presented in [10]. The authors of this paper propose a deep learning method for vehicle detection in bird's eye view using Range-Azimuth-Doppler tensors.…”

Section: Related Workmentioning

confidence: 99%

Weakly Supervised Deep Learning Method for Vulnerable Road User Detection in FMCW Radar

Dimitrievski

Shopovska

Hamme

et al. 2020

2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

Millimeter-wave radar is currently the most effective automotive sensor capable of all-weather perception. In order to detect Vulnerable Road Users (VRUs) in cluttered radar data, it is necessary to model the time-frequency signal patterns of human motion, i.e. the micro-Doppler signature. In this paper we propose a spatio-temporal Convolutional Neural Network (CNN) capable of detecting VRUs in cluttered radar data. The main contribution is a weakly supervised training method which uses abundant, automatically generated labels from camera and lidar for training the model. The input to the network is a tensor of temporally concatenated range-azimuth-Doppler arrays, while the ground truth is an occupancy grid formed by objects detected jointly in-camera images and lidar. Lidar provides accurate ranging ground truth, while camera information helps distinguish between VRUs and background. Experimental evaluation shows that the CNN model has superior detection performance compared to classical techniques. Moreover, the model trained with imperfect, weak supervision labels outperforms the one trained with a limited number of perfect, hand-annotated labels. Finally, the proposed method has excellent scalability due to the low cost of automatic annotation.

show abstract

“…Therefore, improving the perception ability of the vehicles was the broad and active research area so far. Particularly, great number of research works focused on radar-based [21]- [23], lidar-based [24]- [28], and camera-based [29]- [32] object detection.…”

Section: Introductionmentioning

confidence: 99%

REF-Net: Robust, Efficient, and Fast Network for Semantic Segmentation Applications Using Devices With Limited Computational Resources

2021

View full text Add to dashboard Cite

Considering importance of the autonomous driving applications for mobile devices, it is imperative to develop both fast and accurate semantic segmentation models. Thanks to emergence of Deep Learning (DL) techniques, the segmentation models enhanced their accuracy. However, this improved performance of currently popular DL models for self-driving car applications come at the cost of time and computational efficiency. Moreover, networks with efficient model architecture experience lack of accuracy. Therefore, in this study, we propose robust, efficient, and fast network (REF-Net) that combines carefully formulated encoding and decoding paths. Specifically, the contraction path uses mixture of dilated and asymmetric convolution layers with skip connections and bottleneck layers, while the decoding path benefits from nearest neighbor interpolation method that demands no trainable parameters to restore original image size. This model architecture considerably reduces the number of trainable parameters, required memory space, training, and inference time. In fact, the proposed model required nearly 90 times fewer trainable parameters and approximately 4 times less memory space that allowed 3-fold faster training runtime and 2-fold inference speedup in the conducted experiments using Cambridge-driving Labeled Video Database (CamVid) and Cityscapes datasets. Moreover, despite its notable efficiency in terms of memory and time, the REF-Net attained superior results in several segmentation evaluation metrics that showed roughly 2%, 4%, and 3% increase in pixel accuracy, Dice coefficient, and Jaccard Index, respectively. INDEX TERMS Autonomous driving, deep convolutional neural networks, nearest neighbor interpolation, semantic segmentation.

show abstract

Vehicle Detection With Automotive Radar Using Deep Learning on Range-Azimuth-Doppler Tensors

Cited by 154 publications

References 18 publications

300 GHz radar object recognition based on deep neural networks and transfer learning

300 GHz radar object recognition based on deep neural networks and transfer learning

Weakly Supervised Deep Learning Method for Vulnerable Road User Detection in FMCW Radar

REF-Net: Robust, Efficient, and Fast Network for Semantic Segmentation Applications Using Devices With Limited Computational Resources

Contact Info

Product

Resources

About