AI-Enhanced 3D RF Representation Using Low-Cost mmWave Radar

Fang, Shiwei; Nirjon, Shahriar

doi:10.1145/3274783.3275210

Cited by 9 publications

(24 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Frequency Modulated Continuous Wave (FMCW) Millimeter Wave (mmWave) radar sensing has been an active research area in recent years, especially in applications such as person/gesture identification [5], [6], car detection/imaging [1], and environment sensing [2], [3]. Usually Synthetic Aperture Radar (SAR) is used in data collection for high resolution, e.g., [7]- [10].…”

Section: Related Workmentioning

confidence: 99%

“…The deep neural network model proposed in this paper completely replaces the model in the Stage 2 of 3DRIMR, and this new model significantly outperforms the original 3DRIMR. There have been a few recent work on mmWave radar based imaging, mapping, and 3D object reconstruction [1]- [3], [11], [12]. Our work is inspired by their promising research results, and due to the low cost and small form factor of commodity mmWave radar sensors, we plan to develop a simple and fast 3D reconstruction system to be attached in our UAV SLAM system [13] for search and rescue in dangerous environment.…”

Section: Related Workmentioning

confidence: 99%

“…Partically, incorrect ghost points in radar signals can be generated due to multi-path effect. Reference [1]- [3] give more detailed discussion on FMCW mmWave radar sensing.…”

Section: A Fmcw Millimeter Wave Radar Sensing and Imagingmentioning

confidence: 99%

“…However further application of mmWave radar in object imaging and reconstruction is quite difficult due to the characteristics of mmWave radar signals such as low resolution, sparsity, and high noise due to multi-path and specularity. Recent work [1]- [3] attempt to design deep learning systems to generate 2D depth images based on mmWave radar signals. 3DRIMR [4] further introduces a design that generates 3D object shapes based on mmWave radar, but the end results are still not satisfactory.…”

Section: Introductionmentioning

confidence: 99%

“…Yue.Sun001@umb.edu 2 Honggang Zhang is in the Department of Engineering, UMass Boston, USA. Honggang.Zhang@umb.edu 3 Zhuoming Huang is in the Department of Engineering, UMass Boston, USA. Zhuoming.Huang001@umb.edu 4 Benyuan Liu is in the Department of Computer Science, UMass Lowell, USA.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

DeepPoint: A Deep Learning Model for 3D Reconstruction in Point Clouds via mmWave Radar

Sun,

Zhang,

Huang

et al. 2021

Preprint

View full text Add to dashboard Cite

Recent research has shown that mmWave radar sensing is effective for object detection in low visibility environments, which makes it an ideal technique in autonomous navigation systems such as autonomous vehicles. However, due to the characteristics of radar signals such as sparsity, low resolution, specularity, and high noise, it is still quite challenging to reconstruct 3D object shapes via mmWave radar sensing. Built on our recent proposed 3DRIMR (3D Reconstruction and Imaging via mmWave Radar), we introduce in this paper DeepPoint, a deep learning model that generates 3D objects in point cloud format that significantly outperforms the original 3DRIMR design. The model adopts a conditional Generative Adversarial Network (GAN) based deep neural network architecture. It takes as input the 2D depth images of an object generated by 3DRIMR's Stage 1, and outputs smooth and dense 3D point clouds of the object. The model consists of a novel generator network that utilizes a sequence of DeepPoint blocks or layers to extract essential features of the union of multiple rough and sparse input point clouds of an object when observed from various viewpoints, given that those input point clouds may contain many incorrect points due to the imperfect generation process of 3DRIMR's Stage 1. The design of DeepPoint adopts a deep structure to capture the global features of input point clouds, and it relies on an optimally chosen number of DeepPoint blocks and skip connections to achieve performance improvement over the original 3DRIMR design. Our experiments have demonstrated that this model significantly outperforms the original 3DRIMR and other standard techniques in reconstructing 3D objects.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

“…Partically, incorrect ghost points in radar signals can be generated due to multi-path effect. Reference [1]- [3] give more detailed discussion on FMCW mmWave radar sensing.…”

Section: A Fmcw Millimeter Wave Radar Sensing and Imagingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

DeepPoint: A Deep Learning Model for 3D Reconstruction in Point Clouds via mmWave Radar

Sun,

Zhang,

Huang

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

General spiking neural network framework for the learning trajectory from a noisy mmWave radar

Liu

Yan

Deng

et al. 2022

Neuromorph. Comput. Eng.

View full text Add to dashboard Cite

Emerging usages of the millimeter wave (mmWave) radar have drawn extensive attention and inspired exploration of learning mmWave radar data. To be effective, instead of using conventional approaches, recent works have employed modern neural network models to process mmWave radar data. However, due to some inevitable obstacles, e.g., noise and sparsity issues in data, existing approaches are generally customized for specific scenarios. In this paper, we propose a general neuromorphic framework, termed mm-SNN, to process mmWave radar data with spiking neural networks (SNNs), leveraging the intrinsic advantages of SNNs in processing noisy and sparse data. Specifically, we first present the overall design of mm-SNN, which is adaptive and easily expanded for multi-sensor systems. Second, we introduce general and straightforward attention-based improvements into mm-SNN to enhance the data representation, helping promote performance. Moreover, we conduct explorative experiments to certify the robustness and effectiveness of mm-SNN. To the best of our knowledge, mm-SNN is the first SNN-based framework that processes mmWave radar data without using extra modules to alleviate the noise and sparsity issues, and at the same time, achieve considerable performance in the task of trajectory estimation.

show abstract