label-quantized coarse-to-fine learning for remote physiological measurement
Activity counts measure attention, not model quality. Missing counts are unknown. Publication status follows the linked source; a listed venue alone does not confirm peer review.
LQ-rPPG is a remote-PPG model that treats the target as quantized labels and refines from coarse to fine. It comes from Jun Seong Lee, Samyeul Noh, Changki Sung and Hyun Myung and is published in Expert Systems with Applications (DOI). The official repo (GitHub, MIT) ships configs and pretrained checkpoints for intra-dataset runs on MMPD, V4V, UBFC-rPPG, COHFACE and PURE, and cross-dataset runs PURE→UBFC, UBFC→PURE, PURE→MMPD and UBFC→MMPD.
label-quantized coarse-to-fine learning framework (per-stage "LQ" checkpoints 1–5 plus an rPPG head, per repo file names)
Remote photoplethysmography (rPPG) enables non-contact measurement of physiological signals from facial videos, offering strong potential for remote healthcare and daily health monitoring. Driven by this potential, various deep learning-based rPPG methods have been proposed to improve rPPG estimation. However, previous deep learning-based rPPG methods have paid little attention to the quality of training labels and their impact on model learning. Contact-based PPG signals used as training labels often contain noise and variability caused by motion artifacts, inconsistent sensor contact, and morphological distortions. Such label inconsistency can lead models to overfit to the label noise and variability and consequently degrade generalization performance. To address this issue, we propose LQ-rPPG, a label-quantized coarse-to-fine learning framework for robust rPPG estimation. LQ-rPPG consists of a label quantization module and a coarse-to-fine rPPG estimation model. The label quantization module transforms continuous PPG signals into multi-bit quantized pseudo labels with reduced noise and variability. The coarse-to-fine estimation model progressively refines rPPG signals under hierarchical supervision guided by the multi-bit pseudo labels. This design alleviates overfitting to label-specific variations and enables the model to learn structured and consistent representations. As a result, LQ-rPPG achieves robust and generalizable rPPG estimation even under challenging conditions. Experiments on multiple benchmark datasets demonstrate that LQ-rPPG achieves strong performance in both intra- and cross-dataset evaluations, while reducing parameters and multiply-accumulate operations by 88% and 29%, respectively, and increasing throughput by 191%. The code is available at https://github.com/Anonymous-repo-code/LQ-rPPG.
Verbatim fromSemantic Scholar· Abstract via Semantic Scholar; © the authors/publisher
Datasets linked in the hub; pretraining vs evaluation roles follow the paper.
LQ-rPPG. (2026). Expert Systems with Applications , issue dated. https://doi.org/10.1016/j.eswa.2026.132977
@article{lqrppg2026lqrppg,
title = {{LQ-rPPG}},
year = {2026},
journal = {Expert Systems with Applications , issue dated},
publisher = {Elsevier},
doi = {10.1016/j.eswa.2026.132977}
}Record updated
