Models, datasets
& papers.

Name: Human activity recognition Leaderboard
Creator: ChatPPG
License: https://creativecommons.org/licenses/by/4.0/

Explore 162 models, 300 datasets and 580 papers. Start with a model family, narrow by signal, and follow the evidence.

11 benchmark tasks Authors & organizations Graph explorer Download the atlas

Four ways to learn from signals

Explore model families.

All 162 models

Flow previews are paused. Each diagram shows the complete process.

0143 modelsSignal encodersEncode a waveform into features for downstream tasks.Explore models 0222 modelsMultimodalLearn relationships across pulse, heart and motion signals.Explore models 0316 modelsSensor–languageAlign recordings with text for retrieval or interpretation.Explore models 042 modelsForecastingUse past measurements to estimate what may come next.Explore models

Conceptual diagrams, not model architectures. One primary role per record; families can overlap.

Other / unclassified 79

Compare benchmarks

Sort

162 models shown of 162 · loading the rest

Export

Models matching the selected filters
Model	Signals	Model family	Availability	Posted / published	Sources
VarMAE variable-input masked autoencoder for wrist-worn accelerometry Summary & evidence	IMU	Encoders	No open weights	Oct 12, 2026Under review	Paper arXiv
Vision4PPG Vision4PPG turns 1D PPG into 2D images (STFT magnitude, STFT phase, and recurrence plots) and fine-tunes public vision foundation models, DINOv3 and SigLIP-2 ViT-B (86M parameters), with LoRA. It is an adaptation of vision models, not a PPG-pretrained foundation model. In the authors' Table 1, those vision models win 9 of 14 diastolic/systolic MAE comparisons against fully tuned MOMENT-385M and PPG-GPT-345M; SigLIP-2's Aurora-Oscillometric DBP/SBP MAE is 5.32/18.57 mmHg. The bed BCG set is an exception, where MOMENT and PPG-GPT report lower error than either vision model. Summary & evidence	PPG	Other / unclassified	No open weights	2025Preprint	Paper DOI Code
VITAL visual-physiological token alignment for VLM-based rPPG (pointer) Summary & evidence	Not reported	Other / unclassified	Weights unconfirmed	2026Preprint	arXiv
VitalAgent tool-augmented ECG/PPG monitoring agent (Univ. of Melbourne, Ting Dang group) 4 stars Summary & evidence	PPG ECG Text	Sensor–language	No open weights	May 28, 2026Preprint	Paper Code
WatchSleepNet code release Summary & evidence	PPG HRV	Other / unclassified	No open weights	Not reportedUnconfirmed	Code
wav2sleep SOTA ×4 Summary & evidence	PPG ECG Resp EOG	Other / unclassified	Open weights	2024Preprint	Paper Code
WavesFM hierarchical longitudinal FM for raw wearable PPG + accelerometer waveforms Summary & evidence	PPG IMU	Multimodal	Weights unconfirmed	May 9, 2026Preprint	Paper
wHuBERT HuBERT-style Conformer for 25 Hz wearable PPG + accelerometry Summary & evidence	PPG IMU	Multimodal	No open weights	Not reportedUnder review	Paper
Wonderwall Wonderwall is an activity recognition foundation model trained first on physics-based simulated motion and then on curated real datasets. Tested on ten activity recognition benchmarks, it gave state-of-the-art generalization in zero-shot, few-shot and full-shot settings. 3 citations Summary & evidence	IMU	Encoders	No open weights	Mar 16, 2026Published article	Paper
xECG bidirectional xLSTM ECG foundation model (BenchECG) 1 citation 24 stars 475 HF / 30d Summary & evidence	ECG	Encoders	Open weights	Sep 12, 2025Published article	Paper DOI Code
xMAE xMAE is a biosignal pretraining method that reconstructs one signal from another, for example ECG and the delayed PPG pulse it produces, so the model learns physiological timing. Its representations beat unimodal and multimodal baselines on 15 of 19 tasks, including cardiovascular outcomes, abnormal lab tests, sleep staging and demographics, across different devices and body locations. 17 stars Summary & evidence	PPG ECG	Multimodal	No open weights	May 1, 2026Preprint	Paper Code
ZARA code release Summary & evidence	IMU Text	Other / unclassified	No open weights	Not reportedUnconfirmed	Code

Encoders

VarMAE

variable-input masked autoencoder for wrist-worn accelerometry

Summary & evidence

IMU

No open weightsUnder review · Oct 12, 2026

Paper arXiv

Other / unclassified

Vision4PPG

Vision4PPG turns 1D PPG into 2D images (STFT magnitude, STFT phase, and recurrence plots) and fine-tunes public vision foundation models, DINOv3 and SigLIP-2 ViT-B (86M parameters), with LoRA. It is an adaptation of vision models, not a PPG-pretrained foundation model. In the authors' Table 1, those vision models win 9 of 14 diastolic/systolic MAE comparisons against fully tuned MOMENT-385M and PPG-GPT-345M; SigLIP-2's Aurora-Oscillometric DBP/SBP MAE is 5.32/18.57 mmHg. The bed BCG set is an exception, where MOMENT and PPG-GPT report lower error than either vision model.

Summary & evidence

PPG

No open weightsPreprint · 2025

Paper DOI Code

Other / unclassified

VITAL

visual-physiological token alignment for VLM-based rPPG (pointer)

Summary & evidence

Weights unconfirmedPreprint · 2026

arXiv

Sensor–language

VitalAgent

tool-augmented ECG/PPG monitoring agent (Univ. of Melbourne, Ting Dang group)

4 stars

Summary & evidence

PPG ECG Text

No open weightsPreprint · May 28, 2026

Paper Code

Other / unclassified

WatchSleepNet

code release

Summary & evidence

PPG HRV

No open weightsUnconfirmed · Not reported

Summary & evidence

Open weightsPreprint · 2024

Paper Code

Multimodal

WavesFM

hierarchical longitudinal FM for raw wearable PPG + accelerometer waveforms

Summary & evidence

PPG IMU

Weights unconfirmedPreprint · May 9, 2026

Paper

Multimodal

wHuBERT

HuBERT-style Conformer for 25 Hz wearable PPG + accelerometry

Summary & evidence

PPG IMU

No open weightsUnder review · Not reported

Paper

Encoders

Wonderwall

Wonderwall is an activity recognition foundation model trained first on physics-based simulated motion and then on curated real datasets. Tested on ten activity recognition benchmarks, it gave state-of-the-art generalization in zero-shot, few-shot and full-shot settings.

3 citations

Summary & evidence

IMU

No open weightsPublished article · Mar 16, 2026

Paper

Encoders

xECG

bidirectional xLSTM ECG foundation model (BenchECG)

1 citation 24 stars 475 HF / 30d

Summary & evidence

ECG

Open weightsPublished article · Sep 12, 2025

Paper DOI Code

Multimodal

xMAE

xMAE is a biosignal pretraining method that reconstructs one signal from another, for example ECG and the delayed PPG pulse it produces, so the model learns physiological timing. Its representations beat unimodal and multimodal baselines on 15 of 19 tasks, including cardiovascular outcomes, abnormal lab tests, sleep staging and demographics, across different devices and body locations.

17 stars

Summary & evidence

PPG ECG

No open weightsPreprint · May 1, 2026

Paper Code

Other / unclassified

ZARA

code release

Summary & evidence

IMU Text

No open weightsUnconfirmed · Not reported

Code

Showing 151–162 of 162 models

Families describe the main role reported in each record. Signals may describe training data, rather than required inference inputs. Dates use the first arXiv posting when known, otherwise the catalog paper date. Missing information stays unconfirmed. Saves stay in this browser.

Compare benchmark evidence

All benchmarks

Curated collections

Blood pressure datasets Raw wrist PPG recordings PPG foundation models compared Open-weight PPG models Atrial fibrillation datasets Heart rate under motion Stress and mental workload Wearable sleep staging SpO₂ and skin tone Glucose estimation datasets Activity recognition (HAR) datasets

Download the full atlas

Complete source records (JSON and CSV) are free for members. Sign in to download. Every record is readable on this site, and llms.txt describes the hub for AI tools.

For a citable version that never changes, use the 2026-10 snapshot.

Cite or reuse this

CC BY 4.0ChatPPG compiled this data. You can copy, adapt and republish it, including in articles, slides, papers and AI tools, as long as you credit ChatPPG and link to this page. Please also cite the original studies for any number you use.

Attribution

Source: ChatPPG, ChatPPG Research Hub: wearable and PPG benchmark leaderboards, datasets and models, https://chatppg.com/research (CC BY 4.0)

HTML link

Source: <a href="https://chatppg.com/research">ChatPPG, ChatPPG Research Hub: wearable and PPG benchmark leaderboards, datasets and models</a> (<a href="https://creativecommons.org/licenses/by/4.0/" rel="license">CC BY 4.0</a>)

Cite the dated snapshot 2026-10 for a fixed, reproducible version. Please also cite the original publications listed above.

APA

ChatPPG. (2026). ChatPPG Research Hub: wearable and PPG benchmark leaderboards, datasets and models (Version 2026-10) [Data set]. ChatPPG Research Hub. https://chatppg.com/research

BibTeX

@misc{chatppg_research_hub,
  author       = {{ChatPPG}},
  title        = {{ChatPPG Research Hub: wearable and PPG benchmark leaderboards, datasets and models}},
  year         = {2026},
  month        = oct,
  version      = {2026-10},
  howpublished = {\url{https://chatppg.com/research}},
  note         = {ChatPPG Research Hub, Snapshot 2026-10, https://chatppg.com/research/snapshots/2026-10}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this resource, please cite it as below."
type: dataset
title: "ChatPPG Research Hub: wearable and PPG benchmark leaderboards, datasets and models"
authors:
  - name: "ChatPPG"
url: "https://chatppg.com/research"
repository-artifact: "https://chatppg.com/research/snapshots/2026-10"
version: "2026-10"
date-released: "2026-10-17"

Explore model families.

Compare benchmark evidence

Curated collections

Download the full atlas

Cite or reuse this

Attribution

HTML link

APA

Get new PPG datasets, papers and models