Deep Learning

Neural Networks for Human Action Recognition

HAR using multiple sensors including RGB, depth, IMU, and event-based sensors with advanced feature fusion

Jihwan Won, Hanwoong Ryu, Sunwoo Yeon

Deep ECGNet: Ultra Short-Term Stress Monitoring

A deep learning framework that detects mental stress directly from raw, ultra short-term ECG windows.

Deep ECGNet: Ultra Short-Term Stress Monitoring

Deep-ACTINet: End-to-End Sleep/Wake Detection

A wrist-actigraphy deep learning pipeline that beats classical sleep scoring without handcrafted features.

Deep-ACTINet: End-to-End Sleep/Wake Detection

Arrhythmia Classification with Deep Learning

Detecting irregular heartbeats through optimized deep neural architectures for wearable and clinical ECG.

Cheolsoo Park, Yoontae-Park

Bio-signal Analysis with Deep Learning

Applying CNNs and RNNs to ECG, PPG, and multi-sensor streams for real-time healthcare monitoring.

Automatic Image & Data Extraction

Document intelligence pipelines that detect figures, captions, and structured data using deep learning.

Automatic Image & Data Extraction