Learning to see through haze: Radar-based Human Detection for Adverse Weather Conditions

Majer, Filip; Yan, Zhi; Broughton, George; Ruichek, Yassine

doi:10.1109/ecmr.2019.8870954

Cited by 20 publications

(12 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Imaging through scattering and diffusive media has been an important problem for many decades, with numerous solutions reported so far [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. In various fields, including e.g., biomedical optics [5,20], atmospheric physics [6,21], remote sensing [22,23], astronomy [24,25], oceanography [26,27], security [28,29] as well as autonomous systems and robotics [30][31][32], the capability to rapidly see through diffusive and scattering media is of utmost importance. In principle, with a prior information of the transmission matrix of a diffuser [16,33], the distorted images can be recovered using a computer.…”

Section: Main Textmentioning

confidence: 99%

Computational imaging without a computer: seeing through random diffusers at the speed of light

Luo

Zhao

et al. 2022

eLight

117

View full text Add to dashboard Cite

Imaging through diffusers presents a challenging problem with various digital image reconstruction solutions demonstrated to date using computers. Here, we present a computer-free, all-optical image reconstruction method to see through random diffusers at the speed of light. Using deep learning, a set of transmissive diffractive surfaces are trained to all-optically reconstruct images of arbitrary objects that are completely covered by unknown, random phase diffusers. After the training stage, which is a one-time effort, the resulting diffractive surfaces are fabricated and form a passive optical network that is physically positioned between the unknown object and the image plane to all-optically reconstruct the object pattern through an unknown, new phase diffuser. We experimentally demonstrated this concept using coherent THz illumination and all-optically reconstructed objects distorted by unknown, random diffusers, never used during training. Unlike digital methods, all-optical diffractive reconstructions do not require power except for the illumination light. This diffractive solution to see through diffusers can be extended to other wavelengths, and might fuel various applications in biomedical imaging, astronomy, atmospheric sciences, oceanography, security, robotics, autonomous vehicles, among many others.

show abstract

Section: Main Textmentioning

confidence: 99%

Computational imaging without a computer: seeing through random diffusers at the speed of light

Luo

Zhao

et al. 2022

eLight

117

View full text Add to dashboard Cite

show abstract

“…III-D, in fog, rain, and snow, small droplets and snowflakes cause the distortions of LiDAR depth estimation, while there is limited impact on radar sensors. To address this challenge, Majer et al [128] proposed an online method which uses the LiDARbased static detector provides labels for the radar point clouds to train a SVM classifier as a dynamic detector. During adverse weather conditions, a learning-based sensor fusion scheme was designed to make the system more rely on the radar to achieve higher accuracy of pedestrian detection and localization.…”

Section: E Sensor Fusionmentioning

confidence: 99%

“…It can undoubtedly save humans from tedious offline training tasks, and in mobile robotics, this learning paradigm further enables robots to learn on-site in their deployment place to adapt to changes in the environment. This paradigm has been used not only for LiDAR [36], [21], but also for millimeter wave radar with similar data representation [128], [57], not only for indoor service robots [36], but also for autonomous driving perception in urban environments [21].…”

Section: E Sensor Fusionmentioning

confidence: 99%

3D ToF LiDAR in Mobile Robotics: A Review

Yang¹,

Li²,

Zhao³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In the past ten years, the use of 3D Time-of-Flight (ToF) LiDARs in mobile robotics has grown rapidly. Based on our accumulation of relevant research, this article systematically reviews and analyzes the use 3D ToF LiDARs in research and industrial applications. The former includes object detection, robot localization, long-term autonomy, LiDAR data processing under adverse weather conditions, and sensor fusion. The latter encompasses service robots, assisted and autonomous driving, and recent applications performed in response to public health crises. We hope that our efforts can effectively provide readers with relevant references and promote the deployment of existing mature technologies in real-world systems.

show abstract

“…Pioneering work in this field can be traced back more than two decades [19], in which a lifelong learning perspective for mobile robot control is presented. With the rapid development of various related technologies including hardware and algorithms, in recent years, research on robotic online learning has become more and more extensive [13], [20], [14], [15], [21], [22]. In particular, Teichman and Thrun [14] presented a semisupervised learning to the problem of track classification in 3D LiDAR data based on Expectation Maximization (EM) algorithm, which illustrated that learning of dynamic objects can benefit from tracking system.…”

Section: Related Workmentioning

confidence: 99%

Efficient Online Transfer Learning for 3D Object Classification in Autonomous Driving

Yang¹,

Yan²,

Tao³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Autonomous driving has achieved rapid development over the last few decades, including the machine perception as an important issue of it. Although object detection based on conventional cameras has achieved remarkable results in 2D/3D, non-visual sensors such as 3D LiDAR still have incomparable advantages in the accuracy of object position detection. However, the challenge also exists with the difficulty in properly interpreting point cloud generated by LiDAR. This paper presents a multi-modal-based online learning system for 3D LiDAR-based object classification in urban environments, including cars, cyclists and pedestrians. The proposed system aims to effectively transfer the mature detection capabilities based on visual sensors to the new model learning based on non-visual sensors through a multi-target tracker (i.e. using one sensor to train another). In particular, it integrates the Online Random Forests (ORF) [1] method, which inherently has the abilities of fast and multi-class learning. Through experiments, we show that our system is capable of learning a high-performance model for LiDAR-based 3D object classification on-the-fly, which is especially suitable for robotics in-situ deployment while responding to the widespread challenge of insufficient detector generalization capabilities.

show abstract

Learning to see through haze: Radar-based Human Detection for Adverse Weather Conditions

Cited by 20 publications

References 36 publications

Computational imaging without a computer: seeing through random diffusers at the speed of light

Computational imaging without a computer: seeing through random diffusers at the speed of light

3D ToF LiDAR in Mobile Robotics: A Review

Efficient Online Transfer Learning for 3D Object Classification in Autonomous Driving

Contact Info

Product

Resources

About