ParallelEye Pipeline: An Effective Method to Synthesize Images for Improving the Visual Intelligence of Intelligent Vehicles

Li, Xuan; Wang, Kunfeng; Gu, Xianfeng; Deng, Fang; Wang, Fei-Yue

doi:10.1109/tsmc.2023.3273896

Cited by 27 publications

(3 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, a virtual obstacle could learn online to be more "cunning" over time, enhancing its ability to challenge a racing car on the track [34,35]. Also, experienced human Formula One drivers may be invited to remotely manipulate the virtual objects, converting them to rivals on the track alongside the competing racecars [36]. This idea further intensifies the competitive nature of races, pushing both human and artificial intelligence to their limits [37].…”

Section: Enriching Dynamic Events On Racetrackmentioning

confidence: 99%

Toward Fair and Thrilling Autonomous Racing: Governance Rules and Performance Metrics for the Autonomous One

Fang

et al. 2023

IEEE Trans. Intell. Veh.

Self Cite

View full text Add to dashboard Cite

This letter is the second report from a series of IEEE TIV's Decentralized and Hybrid Workshops (DHWs) on Intelligent Vehicles for Education (IV4E). It outlines the prospect of Autonomous One, a future autonomous racing competition modeled after Formula One to advance IV4E by pushing the boundaries of artificial intelligence. Existing autonomous races are facing challenges, such as compromised fairness, low participant enthusiasm, and limited spectator engagement, thereby reducing the educational value of the races. To prevent similar flaws in Autonomous One, our focus lies in setting governance rules from an organizer's perspective. During these DHWs, we analyzed the rules of existing autonomous races before providing rule-making suggestions for Autonomous One. To improve fairness, we recommend a balanced scoring system with rigorous monitoring of participating teams' performance. To discourage consistent victories by a single team, we propose that leading teams share their source codes, thereby setting the championship-level performance as the baseline in the next season and accelerating the evolution of all participating teams' abilities. For enhancing suspense and spectator interest, an on-site adaptive rewarding scheme should be deployed to create thrilling turnarounds. Our strategies aim to maintain fairness, increase spectator interest, inspire competitiveness, and ultimately contribute to the advancement of IV4E. 

show abstract

Section: Enriching Dynamic Events On Racetrackmentioning

confidence: 99%

Toward Fair and Thrilling Autonomous Racing: Governance Rules and Performance Metrics for the Autonomous One

Fang

et al. 2023

IEEE Trans. Intell. Veh.

Self Cite

View full text Add to dashboard Cite

show abstract

“…To address some of these limitations, different non-AI approaches that are based on traditional computer graphics methods may be employed. Synthetic images can be rendered from various angles and under various lighting conditions using 3D models and textures [49,[75][76][77][78][79][80][81]. Images can also be generated using algorithms and rules.…”

Section: Ruler Detection In Imagesmentioning

confidence: 99%

Rulers2023: An Annotated Dataset of Synthetic and Real Images for Ruler Detection Using Deep Learning

Matuzevičius

2023

Electronics

View full text Add to dashboard Cite

This research investigates the usefulness and efficacy of synthetic ruler images for the development of a deep learning-based ruler detection algorithm. Synthetic images offer a compelling alternative to real-world images as data sources in the development and advancement of computer vision systems. This research aims to answer whether using a synthetic dataset of ruler images is sufficient for training an effective ruler detector and to what extent such a detector could benefit from including synthetic images as a data source. The article presents the procedural method for generating synthetic ruler images, describes the methodology for evaluating the synthetic dataset using trained convolutional neural network (CNN)-based ruler detectors, and shares the compiled synthetic and real ruler image datasets. It was found that the synthetic dataset yielded superior results in training the ruler detectors compared with the real image dataset. The results support the utility of synthetic datasets as a viable and advantageous approach to training deep learning models, especially when real-world data collection presents significant logistical challenges. The evidence presented here strongly supports the idea that when carefully generated and used, synthetic data can effectively replace real images in the development of CNN-based detection systems.

show abstract

“…This approach allows for the resolution of modeling, analysis, and experimentation issues within a unified theoretical framework and has found wide application across various domains. For instance, in the field of complex engineering, it has been applied to real-time safety monitoring of visual intelligence [20], transportation systems [21], oil fields [22], metaverse [23], and so on [24][25][26][27]. In agricultural engineering, it has been utilized for monitoring and managing crop cultivation, precision control of crop manufacturing, and fine control of high-value plant species [28].…”

Section: Introductionmentioning

confidence: 99%

ACP-Based Space Systems: Design, Development, and Operation

Cai,

Meng,

Wang

2024

International Journal of Aerospace Engineering

View full text Add to dashboard Cite

In the context of the rapid advancements in space technology and the increasing complexity of space missions, there is a growing need for efficient and effective approaches to tackle the multifaceted challenges faced by space systems. Traditional methods often fall short in providing comprehensive support throughout the entire life cycle of space systems. To address these challenges, this paper presents a novel parallel space system architecture based on ACP (artificial systems, computational experiments, and parallel execution) and explores its applications in the design, development, and operation of space systems. The proposed architecture integrates artificial systems with actual space systems and employs computational experiments to generate extensive sample data. This approach enhances the accuracy of the artificial systems’ model and optimizes the performance of the real systems, facilitating parallel advancements between the two. The design, development, and operation processes of Q-Sat, implemented using the ACP framework, serve as a case study to illustrate the advantages of parallel space systems. Following adjustments made to the discrepancies between parallel systems under the ACP-based space system framework, the accuracy of missing orbit compensation improved by 86.5%, and the 24-hour forecast positional error was reduced by approximately 65 m. Furthermore, this paper discusses future trends, emphasizing the increasing efficiency and reliability of digitized, integrated, and adaptive space systems. The findings contribute to the understanding of parallel space systems and provide valuable insights for further advancements in the field.

show abstract

ParallelEye Pipeline: An Effective Method to Synthesize Images for Improving the Visual Intelligence of Intelligent Vehicles

Cited by 27 publications

References 38 publications

Toward Fair and Thrilling Autonomous Racing: Governance Rules and Performance Metrics for the Autonomous One

Toward Fair and Thrilling Autonomous Racing: Governance Rules and Performance Metrics for the Autonomous One

Rulers2023: An Annotated Dataset of Synthetic and Real Images for Ruler Detection Using Deep Learning

ACP-Based Space Systems: Design, Development, and Operation

Contact Info

Product

Resources

About