A5045621730- Gaze Tracking and Assistive Technology
- Image and Object Detection Techniques
- Advanced Neural Network Applications
- Augmented Reality Applications
- Multimodal Machine Learning Applications
- Smart Agriculture and AI
- Human Pose and Action Recognition
- Date Palm Research Studies
- Domain Adaptation and Few-Shot Learning
- Insect and Arachnid Ecology and Behavior
- Handwritten Text Recognition Techniques
- EEG and Brain-Computer Interfaces
- Infrared Target Detection Methodologies
- Emotion and Mood Recognition
- Virtual Reality Applications and Impacts
- Advanced Measurement and Detection Methods
University of Electronic Science and Technology of China
2024
Jiangxi Agricultural University
2021-2022
Traditional pest detection methods are challenging to use in complex forestry environments due their low accuracy and speed. To address this issue, paper proposes the YOLOv4_MF model. The model utilizes MobileNetv2 as feature extraction block replaces traditional convolution with depth-wise separated reduce parameters. In addition, coordinate attention mechanism was embedded enhance information. A symmetric structure consisting of a three-layer spatial pyramid pool is presented, an improved...
Mirror detection is of great significance for avoiding false recognition reflected objects in computer vision tasks. Existing mirror frameworks usually follow a supervised setting, which relies heavily on high quality labels and suffers from poor generalization. To resolve this, we instead propose the first weakly-supervised framework also provide scribble-based dataset. Specifically, relabel 10,158 images, most have labeled pixel ratio less than 0.01 take only about 8 seconds to label....
With the development of deep learning, synthetic aperture radar (SAR) image ship detection based on convolutional neural network has made significant progress. However, there are two problems. 1) The false alarm rate is high due to complex background and coherent speckle noise interference. 2) For smaller targets, missed prone occur. In this letter, a novel model (MFTF-Net) multi-feature transformation fusion proposed address issues. First, avoid randomness initial point selection influence...
Augmented Reality in Human-Robot Interaction (AR-HRI) boosts user experience. The key challenge is refining interaction methods to minimize discomfort and enhance quality. This AR - HRI study uses Galvanic Skin Respons (GSR) predict improve comfort. User studies tested strategies an environment. A machine learning model, developed from GSR data, predicted comfort levels informed strategy changes. Comfort metrics were visualized every second using Hololens 2, creating system. method improved...