A5100631521- Advanced Neural Network Applications
- Digital Transformation in Industry
- Autonomous Vehicle Technology and Safety
- Robotic Path Planning Algorithms
- Video Surveillance and Tracking Methods
- Industrial Vision Systems and Defect Detection
- Advanced Vision and Imaging
- Blockchain Technology Applications and Security
- Advanced Image and Video Retrieval Techniques
- IoT and Edge/Fog Computing
- Multimodal Machine Learning Applications
- Robotics and Automated Systems
- Image Enhancement Techniques
- Robotics and Sensor-Based Localization
- Reinforcement Learning in Robotics
- Advanced Sensor and Energy Harvesting Materials
- Manufacturing Process and Optimization
- Domain Adaptation and Few-Shot Learning
- Human-Automation Interaction and Safety
- Laser Material Processing Techniques
- Adversarial Robustness in Machine Learning
- Adhesion, Friction, and Surface Interactions
- Generative Adversarial Networks and Image Synthesis
- Human Motion and Animation
- Video Analysis and Summarization
Chinese Academy of Sciences
2017-2025
University of Chinese Academy of Sciences
2018-2025
Beijing Information Science & Technology University
2024-2025
Technology and Engineering Center for Space Utilization
2025
Institute of Automation
2019-2024
Shandong Institute of Automation
2017-2024
National University of Singapore
2023-2024
Beijing Academy of Artificial Intelligence
2022-2024
China Agricultural University
2024
Affiliated Hospital of Jiangxi University of Traditional Chinese Medicine
2024
Very recently, intensive discussions and studies on Industry 5.0 have sprung up caused the attention of researchers, entrepreneurs, policymakers from various sectors around world. However, there is no consensus why what yet. In this paper, we define its philosophical historical origin evolution, emphasize new thinking virtual-real duality human-machine interaction, introduce theory technology based parallel intelligence (PI), artificial societies, computational experiments, execution (the...
Long-tail effect, characterized by highly frequent occurrence of normal scenarios and the scarce appearance extreme "long-tail" scenarios, ubiquitously exists in vision-related problems real-world applications. Though many computer vision methods to date have already achieved feasible performance for most it is still challenging existing systems accurately perceive long-tail scenarios. This deficiency largely hinders practical application systems, since may incur fatal consequences, such as...
The advent of Scenarios Engineering (SE) paves the way to a new era intelligent vehicles (IVs), driven by Artificial Intelligence (AI)-enabled strategies. It aims at shaping IVs be form that is more relevant underlying scenario, thereby accomplishing validation, verification (V&V), and calibration, certification (C&C) each vehicle. However, such improved capabilities relies on accumulation analysis an unprecedented volume scenarios. Recently, Sora other video generation models have opened up...
This perspective paper delves into the concept of foundation intelligence that shapes future smart infrastructure services as transportation sector transitions era Transportation 5.0. First, discussion focuses on a suite emerging technologies essential for intelligence. These encompass digital twinning, parallel intelligence, large vision-language models, traffic simulation and systems modeling, vehicle-to-everything (V2X) connectivity, decentralized/distributed systems. Next, introduces...
installed on the modern intelligent vehicles. Many Artificial Intelligence based foundation models have been proposed for smart sensing to recognize known object classes in new but similar scenarios. However, it is still challenging of detect all both seen and unseen This letter aims at pushing boundary research We first summarize current widely-used intelligence needed then explain Sora-based Parallel Vision boost from basic (1.0) enhanced (2.0) final generalized (3.0). Several...
The rapid development of autonomous driving technology has driven continuous innovation in perception systems, with 4D millimeter-wave (mmWave) radar being one the key sensing devices. Leveraging its all-weather operational characteristics and robust capabilities challenging environments, mmWave plays a crucial role achieving highly automated driving. This review systematically summarizes latest advancements applications field To begin with, we introduce fundamental principles technical...
Video super-resolution has recently become one of the most important mobile-related problems due to rise video communication and streaming services. While many solutions have been proposed for this task, majority them are too computationally expensive run on portable devices with limited hardware resources. To address problem, we introduce first Mobile AI challenge, where target is develop an end-to-end deep learning-based that can achieve a real-time performance mobile GPUs. The...
Cooperative computing is promising to enhance the performance and safety of autonomous vehicles benefiting from increase in amount, diversity as well scope data resources. However, effective privacy-preserving utilization multi-modal multi-source remains an open challenge during construction cooperative mechanisms. Recently, Transformers have demonstrated their potential unified representation features, which provides a new perspective for fusion diverse inputs intelligent vehicles....
One of the most challenging problems in human-machine co-work is gap between human intention and machine's understanding execution. Large Language Models (LLMs) have been showing superior abilities solving such issue. In this paper, we design a universal framework that embeds LLMs as vehicle "Co-Pilot" driving, which can accomplish specific driving tasks with satisfied based on information provided. Meanwhile, utilization workflow defined to handle interaction humans vehicles, memory...
Human drivers instinctively reason with commonsense knowledge to predict hazards in unfamiliar scenarios and understand the intentions of other road users. However, this essential capability is entirely missing from traditional decision-making systems autonomous driving. In response, paper presents DriveLLM, a framework that integrates large language models (LLMs) existing driving stacks. This integration allows for reasoning decision-making. DriveLLM also features unique cyber-physical...
This perspective introduces the concept and framework of knowledge factories with machines for workers to achieve automation Industry 5.0 intelligent industries.
Since Manufacturing 4.0 faces various challenges, including the risks of data leakage and privacy violation, struggle to meet growing demand for personalization, limitations in harnessing human creativity, it has become crucial embark on a transformation toward 5.0. To this end, we propose DeFACT framework parallel manufacturing 5.0, which focuses safe, efficient personalized collaborative production. In DeFACT, different enterprises workers (i.e., digital, robotic biological workers) are...
THE mining sector historically drove the global economy but at expense of severe environmental and health repercussions, posing sustainability challenges [1]–[3]. Recent advancements on artificial intelligence (AI) are revolutionizing through robotic data-driven innovations [4]–[7]. While AI offers industry advantages, it is crucial to acknowledge potential risks associated with its widespread use. Over-reliance may lead a loss human control over operations in future, resulting unpredictable...
Pavement defect detection is of profound significance regarding road safety, so it has been a trending research topic. In the past years, deep learning based methods have turned into key technology, with advantages high accuracy, strong robustness, and adaptability to complex pavement environments. This paper first reviews on image processing 3D imaging. As for image-based techniques, they can be classified three types how label collected data: fully supervised learning, unsupervised other...
Aiming at the problem of path planning algorithm autonomous parade robot in indoor environment, this paper, based on Dijkstra and A∗ algorithm, introduces influence current node's parent node to heuristic function seeks optimal weight optimize algorithm. In MATLAB different scenarios were simulated, compared with non-improved situation terms searching consumption time, cost traversed grid number other indexes. After using more reasonable changing appropriately, poor real-time capacity was...
Virtual simulation testing is becoming indispensable for the intelligence of intelligent vehicles. However, even most advanced software provides rather limited test conditions. In long run, vehicles are expected to work at SAE (Society Automotive Engineers) level 4 or 5. Researchers should make full use virtual scenarios visual algorithms under various imaging this paper, we create realistic artificial scenes simulate self-driving scenarios, and collect a dataset synthetic images from...
Super-Resolution (SR) is a fundamental computer vision task that aims to obtain high-resolution clean image from the given low-resolution counterpart. This paper reviews NTIRE 2021 Challenge on Video Super-Resolution. We present evaluation results two competition tracks as well proposed solutions. Track 1 develop conventional video SR methods focusing restoration quality. 2 assumes more challenging environment with lower frame rates, casting spatio-temporal problem. In each competition, 247...
Briefing: With the rapid development of information technologies such as digital twin, extended reality, and blockchain, hype around “metaverse” is increasing at astronomical speed. However, much attention has been paid to its entertainment social functions. Considering openness interoperability metaverses, market quality inspection promises explosive growth. In this paper, taking advantage we first propose concept Automated Quality Inspection (AutoQI), which performs integrated covering...
Industrial processes are typical cyber–physical–social systems (CPSSs), where the effective management of employees and efficient control machines play important roles. Traditional industries heavily rely on human labor neglect development collection–utilization–transmission integrated information loops, thereby leading to high costs low efficiency in operational procedures. To facilitate natural interactions smart operations for humans machines, industrial foundation models (IFMs) based...
Large-scale synthetic traffic image datasets have been widely used to make compensate for the insufficient data in real world. However, mismatch domain distribution between and hinders application of dataset actual vision system intelligent vehicles. In this paper, we propose a novel synthetic-to-real adaptation method settle from two aspects, i.e., level knowledge level. On level, Style-Content Discriminated Data Recombination (SCD-DR) module is proposed, which decouples style content...
Aiming at improving the enterprise performance and venture capital management, edge computing is combined with blockchain technology in construction of supply chains so as to improve stability security raw material enterprises. Firstly, relationships among capital, performance, chain integration are analyzed, positive impacts optimization innovation on improvement discussed. Secondly, adopted optimize chain, a protocol model established information transmission. In addition, Internet Things...