AnweshCR7/RhythmNet
End-to-end Heart Rate Estimation from Face via Spatial-temporal Representation. A replication of the RhythmNet model.
Implements spatial-temporal feature extraction using a CNN-GRU architecture that processes 300-frame video windows to capture both facial appearance changes and temporal heart rate dynamics. Evaluated on the VIPL-HR dataset with 5-fold cross-validation, achieving ~9.3 bpm MAE without the GRU component. The model operates on raw video input, eliminating the need for explicit preprocessing or ROI detection steps.
No commits in the last 6 months.
Stars
65
Forks
14
Language
Python
License
MIT
Category
Last pushed
Jul 06, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/AnweshCR7/RhythmNet"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
remotebiosensing/rppg
Benchmark Framework for fair evaluation of rPPG
Mobile-Sensing-and-UbiComp-Laboratory/NormWear
A Foundation Model for Multivariate Wearable Sensing of Physiological Signals.
MahdiFarvardin/MEDVSE
Official repository of "Efficient Deep Learning-based Estimation of the Vital Signs on Smartphones".
PhysiologicAILab/FactorizePhys
FactorizePhys: Matrix Factorization for Multidimensional Attention in Remote Physiological...
Sakib1263/NABNet
NABNet: A Nested Attention-guided BiConvLSTM Network for a robust prediction of Blood Pressure...