A5069064265- Medical Image Segmentation Techniques
- Video Surveillance and Tracking Methods
- Vector-borne infectious diseases
- Climate change and permafrost
- Insect Pest Control Strategies
- Plant Parasitism and Resistance
- Human Pose and Action Recognition
- Soft Robotics and Applications
- Autonomous Vehicle Technology and Safety
- Anomaly Detection Techniques and Applications
- Cardiovascular Health and Disease Prevention
- Multimodal Machine Learning Applications
- Cardiovascular Function and Risk Factors
- Surgical Simulation and Training
- Viral Infections and Vectors
- Fire effects on ecosystems
- Malaria Research and Control
- Cryospheric studies and observations
- Gait Recognition and Analysis
- Robotics and Sensor-Based Localization
- Evacuation and Crowd Dynamics
Peking University
2022-2023
Soochow University
2023
National Institute for Parasitic Diseases
2023
Soochow University
2023
University of Electronic Science and Technology of China
2019-2022
Recently, transformer-based methods have been introduced to estimate 3D human pose from multiple views by aggregating the spatial-temporal information of joints achieve lifting 2D 3D. However, previous approaches cannot model inter-frame correspondence each view's joint individually, nor can they directly consider all view interactions at time, leading insufficient learning multi-view associations. To address this issue, we propose a Spatial-View-Temporal transformer (SVTformer) decouple...
In robot-assisted cardiac surgery, predicting heart motion can help improve the operation accuracy and safety of surgical robots. Different from conventional prediction schemes which model point interest (POI) with only temporal correlation past observations, this paper proposes an LSTM-based method by exploiting spatio-temporal 3D movements POI auxiliary points (APs) on same surface heart. Three different LSTM models are investigated. The first two define as a pure time-series forecasting...
Navigating safely and efficiently in dense crowds remains a challenging problem for mobile robots. The interaction mechanisms involved collision avoidance require robots to exhibit active foresighted behaviors while understanding the crowd dynamics. Deep reinforcement learning methods have shown superior performance compared model-based approaches. However, existing lack an intuitive quantitative safety evaluation agents, they may potentially trap agents local optima during training,...
This paper investigates the backscattering properties of vegetation fire based on ground-based scatterometer measurement in combustion period. According to different states during and after fire, respectively, X-band C-band scattering coefficient full polarization (HH, HV, VH, VV ) are measured exploiting spatial temporal responses effect vegetation. Meanwhile, relevant influencing factors, which include parameters vegetation, ground environment, obtained observation area. Besides process...
Abstract The protozoan parasite Babesia microti that causes the zoonoses, babesiosis interacts with host erythrocytes during its life cycle. So far, no effective vaccines are available to prevent infections. In this study, we identified a B. conserved erythrocyte membrane-associated antigen, Bm 8, as high seroreactivity antigen. Bioinformatic and phylogenetic analysis showed protein is among apicomplexan hemoprotozoa, such , Plasmodium Theileria . recombinant 8 (r 8) was obtained by...
The protozoan parasite Babesia microti that causes the zoonotic disease babesiosis resides in erythrocytes of its mammalian host during life-cycle. No effective vaccines are currently available to prevent infections. We previously identified a highly seroactive antigen, named Bm8, as B. conserved erythrocyte membrane-associated by high-throughput protein chip screening. Bioinformatic and phylogenetic analysis showed this is among apicomplexan hemoprotozoa, such members genera Babesia,...
This work aims to predict the 3D coordinates of point interest (POI) on surface beating heart in dynamic minimally invasive surgery, which can improve manoeuvrability cardiac surgical robots and expand their functions. For accurate robust POI motion prediction, a deep learning technique, gated recurrent unit (GRU), is employed learn spatio-temporal (ST) correlation its auxiliary points (APs) from past trajectories. reference, two neural network models that exploit only spatial temporal...