detect-heart-rate-using-face

Origin Code

1. eulerian-remote-heartrate-detection

• using Cascade, Pyramid and heartrate-detection

2. Heart-rate-measurement-using-camera

• using face detector

3. QRS-detection

• using QRS detector

Dataset

Emotion elicitation trials of mahnob-db

• Twenty of the entire videos were used, and were composed of 144980 frames
• The video was cut every 10 seconds and the remaining videos(Less than 10 seconds) were discarded.
• So we conducted the test with a total of 222 data.

Results (using Cascade-Pyramid-eulerian)

• RMSE: 9.6
• MAE: 6.0

Reference

• We test the video(about 8) of smartphone(Galaxy A8) using file of Dlib(fixed_position)-Pyramid-eulerian
• this best Result is following
• RMSE: 8.8
• MAE: 7.6

How to run

git clone https://github.com/FKgk/detect-heart-rate-using-face.git
cd detect-heart-rate-using-face
conda create -n qrs python=3.8.1
conda activate qrs
pip install -r requirements.txt

OR

git clone https://github.com/FKgk/detect-heart-rate-using-face.git
cd detect-heart-rate-using-face
conda env create -f ./environment.yml

• Put your data to data folder
• Select the code in code folder
• You must adjust the input_path in the code
• Run this code

Sub_code folder

• This is real subfiles
• Cut the video per 10 sec
• Change from bdf format to csv format
• Detecting QRS and get heartrate (using EXG2 signal)

Summary

• If you are interested in this topic, I suggest to take a look at the link.