Self-driving Deep Learning System based on Depth Image Based Rendering and LiDAR Point Cloud

Lin, Guo-Han; Chang, Chun-Hsiang; Chung, Ming-Chun; Fan, Yu-Cheng

doi:10.1109/icce-taiwan49838.2020.9258010

Cited by 4 publications

(3 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The experiments performed on the KITTI dataset showed that the proposed PointR-CNN architecture outperforms state-of-the-art methods by only using point cloud as its input data. For self-driving, a deep learning system can use LiDAR point clouds and depth image-based rendering (DIBR) [32]. The DIBR is used to generate parallax map information and obtain the depth image, which is then combined with LiDAR point cloud to repair the objects in the point cloud image.…”

Section: B Deep Learning For Object Detection 1) Supervised Learningmentioning

confidence: 99%

Autonomous Vehicles Perception (AVP) Using Deep Learning: Modeling, Assessment, and Challenges

Jebamikyous

Kashef

2022

IEEE Access

View full text Add to dashboard Cite

Perception is the fundamental task of any autonomous driving system, which gathers all the necessary information about the surrounding environment of the moving vehicle. The decision-making system takes the perception data as input and makes the optimum decision given that scenario, which maximizes the safety of the passengers. This paper surveyed recent literature on autonomous vehicle perception (AVP) by focusing on two primary tasks: Semantic Segmentation and Object Detection. Both tasks play an important role as a vital component of the vehicle's navigation system. A comprehensive overview of deep learning for perception and its decision-making process based on images and LiDAR point clouds is discussed. We discussed the sensors, benchmark datasets, and simulation tools widely used in semantic segmentation and object detection tasks, especially for autonomous driving. This paper acts as a road map for current and future research in AVP, focusing on models, assessment, and challenges in the field.

show abstract

Section: B Deep Learning For Object Detection 1) Supervised Learningmentioning

confidence: 99%

Autonomous Vehicles Perception (AVP) Using Deep Learning: Modeling, Assessment, and Challenges

Jebamikyous

Kashef

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The use of Light Detection and Ranging (LiDAR) devices spans across decades of history in various implementations for the military, private sector, and even civilian hobbyists (McManamon, Kamerman and Huffaker, 2010;Molebny, Kamerman and Steinvall, 2010). Although the basic principle of LiDAR relies on simply emitting infrared light at a surface or object and capturing what reflects back, the long-term development of LiDAR has produced countless LiDAR enabled technologies that facilitate a range of functions such as personnel detection through foliage (Tether, 2004), ecological measurement (Eitel et al, 2016), self-driving cars (Lin et al, 2020), and digital recreation of 3-dimensional (3D) objects (Raj et al, 2020). However, the technological accomplishments enabled by LiDAR devices often come with a price tag beyond typical consumer budgets due to the specialized nature of LiDAR hardware components and supporting hardware/software (Queralta et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Analysis of Image Thresholding Algorithms for Automated Machine Learning Training Data Generation

Creek

Mullins

2022

iccws

View full text Add to dashboard Cite

Secured compounds often safeguard physical layout details of both internal and external facilities, but these details are at risk due to the growing inclusion of Light Detection and Ranging (LiDAR) sensors in consumer off-the-shelf (COTS) technology such as cell phones. The ability to record detailed distance data with cell phones facilitates the production of high-quality three-dimensional scans in a discrete manner which directly threatens the security of private compounds. Therefore, it behooves the organizations in charge of private compounds to detect LiDAR activity. Many security cameras already detect LiDAR sources as generic light sources in specific conditions, but further analysis must identify these light sources as LiDAR sources in order to alert an organization of a potential security incident. Testing confirms the feasibility of identifying some LiDAR sources based on the color and intensity of light shined directly into a camera sensor, but this analysis proves inadequate for cell phone LiDAR. However, the unique intensity and pattern characteristics of cell phone LiDAR reflected off a surface can potentially be identified by an object identification machine learning model. In order to train a model to identify a LiDAR object, we must first produce a training dataset containing marked and labelled LiDAR objects. To do this, we apply an image thresholding algorithm to isolate the LiDAR object in an image to calculate its bounding box. The image thresholding algorithm directly affects the bounding box accuracy, so we test two different algorithms and find that Otsu’s image thresholding algorithm performs best, resulting in 99.5% accurate bounding boxes.

show abstract

“…Those spatial features are combined with global semantic features for accurate confidence prediction and box refinement. The experiments performed on the KITTI dataset showed that the proposed PointRCNN architecture outperforms state-of-the-art methods by only using point cloud as its input data.For self-driving, a deep learning system can use LiDAR point clouds and depth image-based rendering (DIBR) for self-driving[29]. The DIBR is used to generate parallax map information and obtain the depth image, which is then combined with LiDAR point cloud to repair the objects in the point cloud image.…”

mentioning

confidence: 99%

Deep Learning-Based Semantic Segmentation in Autonomous Driving

Jebamikyous

2024

Preprint

View full text Add to dashboard Cite

<p>Perception is a fundamental task of autonomous driving systems, which gathers all the necessary information about the surrounding environment of the moving vehicle. Then a decision-making system takes the perception data as input and provides the optimum decision given a scenario, which maximizes the safety of the passengers. In this project, we have developed variants of the U-Net model to perform semantic segmentation on urban scene images to understand the surroundings of an autonomous vehicle. The U-Net model and its variants are adopted for semantic segmentation in this project to account for the power of the U-Net in handling large and small datasets. We have also compared the best-performing variant with other commonly used semantic segmentation models. The comparative analysis was performed using three well-known models, including FCN-16, FCN-8, and SegNet. After conducting sensitivity and comparative analysis, it is concluded that the U-Net variants performed the best in terms of the Intersection over Union (IoU) evaluation metric and other quality metrics.</p>

show abstract

Self-driving Deep Learning System based on Depth Image Based Rendering and LiDAR Point Cloud

Cited by 4 publications

References 2 publications

Autonomous Vehicles Perception (AVP) Using Deep Learning: Modeling, Assessment, and Challenges

Autonomous Vehicles Perception (AVP) Using Deep Learning: Modeling, Assessment, and Challenges

Analysis of Image Thresholding Algorithms for Automated Machine Learning Training Data Generation

Deep Learning-Based Semantic Segmentation in Autonomous Driving

Contact Info

Product

Resources

About