imu-sensors

Build your own motion capture system without breaking the bank! 🎮 This project combines BNO055 sensors with ESP32 boards to track movement in real-time. Visualize motion in 3D, stream data wirelessly, and customize for your needs. Great for indie animation, hobbyist VR projects, student research, or just having fun with motion tracking.

Безразметочная кластеризация физической активности по IMU-сенсорам: 534k строк, оконная агрегация, PCA + MiniBatchKMeans. 16 место на Kaggle

An embedded IMU sensing pipeline for FRC robotics integrating ESP32-based sensor fusion with RoboRIO via SPI, employing Kalman filtering, OLS drift estimation, ZUPT stabilization, and adaptive temperature compensation to achieve robust, low-drift orientation tracking in high-noise environments.

Official implementation of “Adapting Pretrained Large Vision Models for Sensor-based Activity Recognition”

Official implementation of "Next Token Prediction for Generalizable Human Activity Recognition"

Machine learning-based human activity recognition using wearable IMU sensors for elderly monitoring. Includes Random Forest, SVM, sensor ablation analysis, and LOSO validation.

Modular time-series pipeline for 6-channel IMU gesture classification using classical ML, 1D CNNs, MongoDB, and user-independent cross-validation.

Active tremor-suppression robotic spoon & stabilizer for Parkinson's patients utilizing closed-loop PID control and 6-DOF IMU filtering.

Indoor fitness action recognition from IMU and camera (MediaPipe pose) data, with a mini program client.