A5081861180- Human Pose and Action Recognition
- Advanced Vision and Imaging
- Emotion and Mood Recognition
- Human Motion and Animation
- 3D Shape Modeling and Analysis
- Digital Mental Health Interventions
- Robotics and Sensor-Based Localization
- Mental Health via Writing
- Bat Biology and Ecology Studies
- Sentiment Analysis and Opinion Mining
- Mental Health Research Topics
- Simulation and Modeling Applications
- Optical measurement and interference techniques
- Traffic control and management
- Gait Recognition and Analysis
- Video Surveillance and Tracking Methods
- Reinforcement Learning in Robotics
- Robot Manipulation and Learning
- Traffic Prediction and Management Techniques
Sichuan Normal University
2025
Qilu University of Technology
2023-2024
Shandong Academy of Sciences
2023-2024
Shandong University
2023-2024
Tsinghua University
2021-2024
Motion capture is facing some new possibilities brought by the inertial sensing technologies which do not suffer from occlusion or wide-range recordings as vision-based solutions do. However, recorded signals are sparse and quite noisy, online performance global translation estimation turn out to be two key difficulties. In this paper, we present TransPose, a DNN-based approach perform full motion (with both translations body poses) only 6 Inertial Measurement Units (IMUs) at over 90 fps....
Human and environment sensing are two important topics in Computer Vision Graphics. motion is often captured by inertial sensors, while the mostly reconstructed using cameras. We integrate techniques together EgoLocate, a system that simultaneously performs human capture (mocap), localization, mapping real time from sparse body-mounted including 6 measurement units (IMUs) monocular phone camera. On one hand, mocap suffers large translation drift due to lack of global positioning signal....
Motion capture is facing some new possibilities brought by the inertial sensing technologies which do not suffer from occlusion or wide-range recordings as vision-based solutions do. However, recorded signals are sparse and quite noisy, online performance global translation estimation turn out to be two key difficulties. In this paper, we present TransPose, a DNN-based approach perform full motion (with both translations body poses) only 6 Inertial Measurement Units (IMUs) at over 90 fps....
Either RGB images or inertial signals have been used for the task of motion capture (mocap), but combining them together is a new and interesting topic. We believe that combination complementary able to solve inherent difficulties using one modality input, including occlusions, extreme lighting/texture, out-of-view visual mocap global drifts mocap. To this end, we propose method fuses monocular sparse IMUs real-time human capture. Our contains dual coordinate strategy fully explore IMU with...
Existing inertial motion capture techniques use the human root coordinate frame to estimate local poses and treat it as an by default. We argue that when has linear acceleration or rotation, should be considered non-inertial theoretically. In this paper, we model fictitious forces are non-neglectable in a auto-regressive estimator delicately designed following physics. With forces, force-related IMU measurement (accelerations) can correctly compensated thus Newton's laws of satisfied. case,...
The giant panda, a rare and iconic species endemic to China, has attracted significant attention from both domestic international researchers due its crucial ecological role, unique cultural value, distinct evolutionary history. While substantial progress been made in the field of individual identification, behavior recognition remains underdeveloped, facing challenges such as lack dynamic temporal features insufficient extraction behavioral characteristics. To address these challenges, we...
Existing inertial motion capture techniques use the human root coordinate frame to estimate local poses and treat it as an by default. We argue that when has linear acceleration or rotation, should be considered non-inertial theoretically. In this paper, we model fictitious forces are non-neglectable in a auto-regressive estimator delicately designed following physics. With forces, force-related IMU measurement (accelerations) can correctly compensated thus Newton's laws of satisfied. case,...
With the rise in societal pressures, depression and anxiety have increasingly become prominent mental health conditions impacting people's lives. To enhance efficacy of automatic detection for these disorders, we developed an experimental framework called Voluntary Facial Expression Mimicry(VFEM). This led to creation VFEM Dataset, which supports related research endeavors. Subsequently, introduce LI-FPN designed specifically identification disorders. The comprises two core components:...
Motion capture is facing some new possibilities brought by the inertial sensing technologies which do not suffer from occlusion or wide-range recordings as vision-based solutions do. However, recorded signals are sparse and quite noisy, online performance global translation estimation turn out to be two key difficulties. In this paper, we present TransPose, a DNN-based approach perform full motion (with both translations body poses) only 6 Inertial Measurement Units (IMUs) at over 90 fps....