A5102980965- Video Surveillance and Tracking Methods
- Human Pose and Action Recognition
- Infrared Target Detection Methodologies
- Advanced Measurement and Detection Methods
- UAV Applications and Optimization
- Fire Detection and Safety Systems
- Advanced Image and Video Retrieval Techniques
- Image Enhancement Techniques
- Advanced Neural Network Applications
- Advanced Vision and Imaging
- AI in cancer detection
- Advanced Algorithms and Applications
- Drilling and Well Engineering
- Retinal Imaging and Analysis
- Machine Learning in Materials Science
- Pulsed Power Technology Applications
- Remote-Sensing Image Classification
- Hydrocarbon exploration and reservoir analysis
- Medical Image Segmentation Techniques
- Mineral Processing and Grinding
- Visual Attention and Saliency Detection
- Robotics and Sensor-Based Localization
- Electromagnetic Launch and Propulsion Technology
Xi'an Jiaotong University
2019-2024
State Key Laboratory of Electrical Insulation and Power Equipment
2024
The Vision Meets Drone (VisDrone2019) Single Object Tracking challenge is the second annual research activity focusing on evaluating single-object tracking algorithms drones, held in conjunction with International Conference Computer (ICCV 2019). VisDrone-SOT2019 Challenge goes beyond its VisDrone-SOT2018 predecessor by introducing 25 more challenging sequences for long-term tracking. We evaluate and discuss results of 22 participating 19 state-of-the-art trackers collected dataset. are...
Object tracking has been studied for decades, but most of the existing works are focused on RGB tracking. For an infrared video, object is often textureless, especially far-range drone planar targets. Furthermore, motion camera and unexpected movement drones make more difficult, causing algorithms lose In this paper a robust realtime algorithm proposed drones, in which feature attention module expansion strategy searching target added to fully convolutional classifier. Experiments Anti-UAV...
Object tracking has been studied for decades, but most of the existing works are focused on short-term tracking. For a long sequence, object is often fully occluded or out view time, and algorithms lose target, it difficult to re-catch target even if reappears again. In this paper novel long-term algorithm flow_MDNet_RPN proposed, in which result judgement module detection added algorithm. Experiments show that proposed effective problem disappearance.
Single-frame infrared small target (SIRST) detection aims to recognize targets from clutter backgrounds. Recently, convolutional neural networks have achieved significant advantages in general object detection. With the development of Transformer, scale SIRST models is constantly increasing. Due limited training samples, performance has not been improved accordingly. The quality, quantity, and diversity dataset are critical targets. To highlight this issue, we propose a negative sample...
In view of the large amount calculation and long operation cycle in particle filter tracking algorithm. This paper puts forward a designed FPGA platform, using fast speed parallel processing mechanism to improve performance system. The experimental results show that system has good real-time accuracy. Its portability is strong, having broad prospect application.