A5063781430- Autonomous Vehicle Technology and Safety
- Robot Manipulation and Learning
- Robotics and Sensor-Based Localization
- Real-Time Systems Scheduling
- Robotic Path Planning Algorithms
- Advanced Neural Network Applications
- Soft Robotics and Applications
- Embedded Systems Design Techniques
- Traffic control and management
- Video Surveillance and Tracking Methods
- Advanced Vision and Imaging
- Reinforcement Learning in Robotics
- Advanced Memory and Neural Computing
- Parallel Computing and Optimization Techniques
- Neural dynamics and brain function
- Surgical Simulation and Training
- Traffic Prediction and Management Techniques
- Advanced Software Engineering Methodologies
- Human Pose and Action Recognition
- Transportation Planning and Optimization
- Formal Methods in Verification
- Anomaly Detection Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Modular Robots and Swarm Intelligence
- Domain Adaptation and Few-Shot Learning
Technical University of Munich
2016-2025
Inform (Germany)
2024
Technische Hochschule Ingolstadt
2023-2024
Bayer (Germany)
2010-2024
University of California, Los Angeles
2023
Fraunhofer Institute for Transportation and Infrastructure Systems
2023
City University of Macau
2023
University of Macau
2023
Eye & ENT Hospital of Fudan University
2023
Northwestern Polytechnical University
2023
Gradient boosting machines are a family of powerful machine-learning techniques that have shown considerable success in wide range practical applications. They highly customizable to the particular needs application, like being learned with respect different loss functions. This article gives tutorial introduction into methodology gradient methods. A theoretical information is complemented many descriptive examples and illustrations which cover all stages model design. Considerations on...
As a bio-inspired and emerging sensor, an event-based neuromorphic vision sensor has different working principle compared to the standard frame-based cameras, which leads promising properties of low energy consumption, latency, high dynamic range (HDR), temporal resolution. It poses paradigm shift sense perceive environment by capturing local pixel-level light intensity changes producing asynchronous event streams. Advanced technologies for visual sensing system autonomous vehicles from...
Pedestrian motion prediction is a fundamental task for autonomous robots and vehicles to operate safely. In recent years many complex approaches based on neural networks have been proposed address this problem. work we show that - surprisingly simple Constant Velocity Model can outperform even state-of-the-art models. This indicates either are not able make use of the additional information they provided with, or as relevant commonly believed. Therefore, analyze how process their input it...
Although great progress has been made in generic object detection by advanced deep learning techniques, detecting small objects from images is still a difficult and challenging problem the field of computer vision due to limited size, less appearance, geometry cues, lack large-scale datasets targets. Improving performance wider significance many real-world applications, such as self-driving cars, unmanned aerial vehicles, robotics. In this article, first-ever survey recent studies...
Ethernet-based protocols are getting more and important for Industry 4.0 the Internet of Things. In this paper, we compare features, package overhead, performance some most in area. First, present a general feature comparison OPC UA, ROS, DDS, MQTT, followed by detailed wire protocol evaluation, which gives an overview over overhead establishing connection sending data. tests evaluate open-source implementations these measuring round trip time messages different system states: idle, high CPU...
Abstract Biologically inspired spiking neural networks are increasingly popular in the field of artificial intelligence due to their ability solve complex problems while being power efficient. They do so by leveraging timing discrete spikes as main information carrier. Though, industrial applications still lacking, partially because question how encode incoming data into spike events cannot be uniformly answered. In this paper, we summarise signal encoding schemes presented literature and...
Image restoration aims to reconstruct a sharp image from its degraded counterpart, which plays an important role in many fields. Recently, Transformer models have achieved promising performance on various tasks. However, their quadratic complexity remains intractable issue for practical applications. The aim of this study is develop efficient and effective framework restoration. Inspired by the fact that different regions corrupted always undergo degradations degrees, we propose focus more...
Image restoration aims to reconstruct the latent sharp image from its corrupted counterpart. Besides dealing with this long-standing task in spatial domain, a few approaches seek solutions frequency domain by considering large discrepancy between spectra of sharp/degraded pairs. However, these algorithms commonly utilize transformation tools, e.g., wavelet transform, split features into several parts, which is not flexible enough select most informative component recover. In paper, we...
Image restoration aims to reconstruct a high-quality image from degraded low-quality observation. Recently, Transformer models have achieved promising performance on tasks due their powerful ability model long-range dependencies. However, the quadratically growing complexity with respect input size makes them inapplicable practical applications. In this paper, we develop an efficient convolutional network for by enhancing multi-scale representation learning. To end, propose omni-kernel...
Image restoration aims to reconstruct a high-quality image from its corrupted version, playing essential roles in many scenarios.Recent years have witnessed paradigm shift convolutional neural networks (CNNs) Transformerbased models due their powerful ability model long-range pixel interactions.In this paper, we explore the potential of CNNs for and show that proposed simple network architecture, termed ConvIR, can perform on par with or better than Transformer counterparts.By re-examing...
Abstract Highly sensitive and humidity‐resistive detection of the most common physical stimuli is primary importance for practical application in real‐time monitoring. Here, a simple yet effective strategy reported to achieve highly humidity‐stable hybrid composite that enables simultaneous accurate pressure temperature sensing single sensor. The improved electronic performance due enhanced planarity poly (3,‐4ethylenedioxythiophene) (PEDOT) charge transfer between PEDOT:polystyrene...
Abstract In recent years, brain research has indisputably entered a new epoch, driven by substantial methodological advances and digitally enabled data integration modelling at multiple scales—from molecules to the whole brain. Major are emerging intersection of neuroscience with technology computing. This science combines high-quality research, across scales, culture multidisciplinary large-scale collaboration, translation into applications. As pioneered in Europe’s Human Brain Project...
In many future joint-action scenarios, humans and robots will have to interact physically in order successfully cooperate. Ideally, seamless human-robot interaction should not require training for the human, but be intuitively simple. Nonetheless, cooperation involve some degree of learning adaptation. Here, we report on a simple case physical interaction, hand-over task. Even such basic task as manually handing over an object from one agent another requires that both partners agree upon...
Tying suture knots is a time-consuming task performed frequently during minimally invasive surgery (MIS). Automating this could greatly reduce total time for patients. Current solutions to problem replay manually programmed trajectories, but more general and robust approach use supervised machine learning smooth surgeon-given training trajectories generalize from them. Since knot tying generally requires controller with internal memory distinguish between identical inputs that require...
Most stereo correspondence algorithms match support windows at integer-valued disparities and assume a constant disparity value within the window. The recently proposed Patch Match algorithm by Bleyer et al. overcomes this limitation of previous directly estimating planes. This work presents method that integrates into variational smoothing formulation using quadratic relaxation. resulting allows explicit regularization normal gradients estimated plane parameters. Evaluation our in...
The Human Brain Project (HBP) is a candidate project in the European Union's FET Flagship Program, funded by ICT Program Seventh Framework Program. will develop new integrated strategy for understanding human brain and novel research platform that integrate all data knowledge we can acquire about structure function of use it to build unifying models be validated simulations running on supercomputers. drive development supercomputing life sciences, generate neuroscientific as benchmark...
Energy optimization is a critical design concern for embedded systems. Combining D VFS +D PM considered as one preferable technique to reduce energy consumption. There have been optimal algorithms periodic independent tasks running on uniprocessor in the literature. Optimal combination of and dependent multicore systems however not yet reported. The challenge this problem that idle intervals cores are easy model. In article, novel proposed directly model individual such both can be optimized...
Ridesharing offers the opportunity to make more efficient use of vehicles while preserving benefits individual mobility. Presenting ridesharing as a viable option for commuters, however, requires minimizing certain inconvenience factors. One these factors includes detours which result from picking up and dropping off additional passengers. This paper proposes method aims best utilize potential keeping below specific limit. The specifically targets systems on very large scale with high degree...