A5054222924- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- UAV Applications and Optimization
- Infrared Target Detection Methodologies
- Visual Attention and Saliency Detection
- Underwater Vehicles and Communication Systems
- Innovation in Digital Healthcare Systems
- Fluid Dynamics Simulations and Interactions
- Soft Robotics and Applications
Changsha University of Science and Technology
2023
Shanghai Maritime University
2023
Beijing Jiaotong University
2023
Moving object segmentation (MOS), aiming at segmenting moving objects from video frames, is an important and challenging task in computer vision with various applications. With the development of deep learning (DL), MOS has also entered era models toward spatiotemporal feature learning. This paper aims to provide latest review recent DL-based methods proposed during past three years. Specifically, we present a more up-to-date categorization based on model characteristics, then compare...
Multi-Object Tracking (MOT) is a key technology for Unmanned Aerial Vehicles (UAVs). Traditional tracking-by-detection methods firstly employ an object detector to retrieve targets in each image and then track them based on matching algorithm. Recently, the popular multi-task learning has been dominating this area since they can detect extract Re-Identification (Re-ID) features computationally efficient way. However, detection task tracking have conflicting requirements features, leading...
As a result of increasing urbanization, traffic monitoring in cities has become challenging task. The use Unmanned Aerial Vehicles (UAVs) provides an attractive solution to this problem. Multi-Object Tracking (MOT) for UAVs is key technology fulfill Traditional detection-based-tracking (DBT) methods begin by employing object detector retrieve targets each image and then track them based on matching algorithm. Recently, the popular multi-task learning have been dominating area, since they can...
Water-jet propulsion helps aquatic creatures achieve both accelerated motion (e.g., flying squid) and long-endurance cruising jellyfish). That inspired the development of unmanned aerial vehicle (AquaUAV) underwater soft robots. Underwater robots water-jet in all directions but have little thrust, AquaUAV has large thrust can only accelerate from to air. Here, improve their deficiencies, we proposed a novel thrusters using cavity-membrane-based (CM-jet) structure. To quantitatively test...
With the development of multi-tasking learning, multi-object tracking (MOT) has been dominated by model Joint Detection and Re-Identification (JDR). FairMOT is a milestone work under such paradigm. It contends that anchor-free methods are more suitable than anchor-based ones for building JDR models. However, detectors widely used in unmanned aerial vehicle (UAV) applications since they tend to achieve higher performance detecting small targets which universal drone videos. In this paper, we...