McJackTang / rppg

implement remote-ppg (rppg;fppg) & cNIBP model using pytorch

Remote Biosensing

---

Our community is eagerly waiting for researchers and developers interested in non-contact/non-invasive algorithm research and development to join us.

Remote Biosensing (rPPG) is a framework for non-contact algorithms for remote photoplethysmography (rPPG) and for non-invasive blood pressure measurement algorithms (CNIBP) technology. We aim to implement a deep learning-based remote photoplethysmography (rPPG) model and continuous non-invasive blood pressure (CNIBP) using PyTorch.

Quick Start with our examples

• rPPG( remote PPG) models

year	type	model	example	config	paper
2018	DL	DeepPhys	example	config	paper
2020	DL	MTTS	example	config	paper
2020	DL	MetaPhys	example	config	paper
2021	DL	EfficentPhys			paper
2023	DL	BIGSMALL			paper
2019	DL	STVEN_rPPGNET			paper
2019	DL	PhysNet	example	config	paper
2019	DL	2D PhysNet + LSTM	example	config	paper
2022	DL	PhysFormer			paper
2023	DL	PhysFormer++			paper
2022	DL	APNET	example	config	paper
TBD	DL	APNETv2	example	config	paper
2019	DL	RhythmNet			paper
2022	DL	JAMSNet			paper
2023	DL	CRGB rPPG			paper
2008	TR	GREEN			paper
2010	TR	ICA			paper
2011	TR	PCA			paper
2013	TR	CHROM			paper
2014	TR	PBV			paper
2016	TR	POS			paper
2015	TR	SSR			paper
2018	TR	LGI			paper
2023	TR	EEMD + FastICA			paper

• CNIBP (Continuous non-invasive blood pressure)

• PP-Net exmaple paper

datasets

You can find information about datasets at the following link.

Documentation(TBD)

Performance Comparison

- rPPG

• All evaluations are based on the model with the lowest loss value during validation.

MODEL	Train/val Dataset	Test Dataset	lr	optim	lr-sch	HR - MAE	HR - RMSE	HR - MAPE	HR -corr
PhysNet	UBFC	PURE	1e-3	Adam	None	1.74	8.40	1.75	0.92
PhysNet	PURE	UBFC	1e-3	Adam	None	1.90	7.02	2.11	0.87

• CNIBP

Bench Mark Git

Community

Our community is eagerly waiting for researchers and developers interested in non-contact/non-invasive algorithm research and development to join us.

Contacts

• Dae Yeol Kim, [e-mail]
• Kwangkee Lee, [e-mail]

Funding

This work was partly supported by the ICT R&D program of MSIP/IITP. [2021(2021-0-00900), Adaptive Federated Learning in Dynamic Heterogeneous Environment]

Reference

If you use this code before our paper is published, please cite the GitHub link. https://github.com/remotebiosensing/rppg