Description: Development of mmPose-FK, a novel mmWave radar-based pose estimation method using dynamic forward kinematics (FK) to overcome challenges like low resolution and noise. Achieves stable joint tracking via deep learning integration.

Key Papers:

1. mmPose-FK: A Forward Kinematics Approach to Dynamic Skeletal Pose Estimation
   S. Hu, S. Cao et al.
   IEEE Sensors Journal, 2024.
2. mmPose-NLP: A Natural Language Processing Approach
   A. Sengupta, S. Cao
   IEEE Transactions on Neural Networks and Learning Systems, 2023.
3. Real-Time Pose Estimation with CNNs
   A. Sengupta et al.
   IEEE Sensor Journal, 2020.

Description: A privacy-preserving system using mmWave radar and deep learning (CNNs, RNNs) to detect falls in real time. Addresses challenges like obstructions and data scarcity.

Key Papers:

1. Radar-Based Fall Detection: A Survey
   S. Hu, S. Cao et al.
   IEEE Robotics and Automation Magazine, 2024.
2. mmFall: Fall Detection Using Hybrid Variational RNN AutoEncoder
   F. Jin et al.
   IEEE Transactions on Automation Science and Engineering, 2020.

Description: Flexible extrinsic calibration of 3D radar and camera using a single corner reflector. Solves PnP with RANSAC and LM optimization.

Key Paper:

• 3D Radar-Camera Co-Calibration
  L. Cheng et al.
  IEEE Radar Conference, 2023.

Description: Online calibration via deep learning to extract common features from radar (Range-Doppler-Angle) and camera data.

Specifically, the extracted common feature serves as an example to demonstrate an online targetless calibration method between the radar and camera systems. The estimation of the extrinsic transformation matrix is achieved through this feature-based approach. To enhance the accuracy and robustness of the calibration, we apply the RANSAC and Levenberg-Marquardt (LM) nonlinear optimization algorithm for deriving the matrix. Additionally, we incorporate adaptive variance measures to ensure efficiency during the optimization process.

Key Paper:

• Online Targetless Calibration Using Common Features
  L. Cheng, S. Cao
  IEEE National Aerospace and Electronics Conference, 2023.

Description: Fusion of radar and camera data using Bi-directional LSTM and tri-output mechanisms for robust tracking.

Key Paper:

• Robust Multi-Object Tracking via Radar-Camera Fusion
  L. Cheng et al.
  IEEE Transactions on Intelligent Transportation Systems, 2024.

Description: Decision-level fusion with tri-Kalman filters for localization accuracy and robustness.

Key Paper:

• Robust Tracking Using Radar-Camera Fusion
  A. Sengupta et al.
  IEEE Sensors Letters, 2022.

Description: Adaptive noise canceller for FMCW radar to improve SIR and reduce ghost targets.

Key Paper:

• Interference Mitigation Using Adaptive Noise Canceller
  F. Jin, S. Cao
  IEEE Transactions on Vehicular Technology, 2019.

Description: Real-time multi-patient behavior detection using mmWave radar and CNNs.

Key Papers:

1. Real-Time Behavior Detection
   R. Zhang, S. Cao
   IEEE Sensors Letters, 2019.
2. Multi-Patient Detection in Real-Time
   F. Jin et al.
   IEEE Radar Conference, 2019.

Application:

• SeVA Technology Demo

Description: Portable 3D imaging using inverse Radon transform on mmWave radar data.

Key Papers:

1. Compressed Sensing for 3D Imaging
   R. Zhang, S. Cao
   IEEE Radar Conference, 2017.
2. Portable mmWave 3D Imaging
   R. Zhang, S. Cao
   IEEE Radar Conference, 2017.