Human activity recognition leaderboard

DEO: balanced accuracy

1 reported result for human activity recognition on DEO, ranked by balanced accuracy. Every row cites its source publication.

This page ranks every result ChatPPG has recorded for human activity recognition on DEO, measured by Balanced accuracy (HAR / class-imbalance) (higher is better). Each number is copied from the publication cited in its row, and ranks apply only within this dataset and metric.

Only one result has been reported on this dataset and metric so far, so this is a reference point rather than a multi-method comparison.

Human activity recognition (IMU): DEO, Balanced accuracy (HAR / class-imbalance). Reported rankings; evaluation protocols may differ.
RankMethod & evaluationBalanced accuracy (HAR / class-imbalance)↑ Higher is better
1Diffusion-Based Synthetic Data Pretraining for E

Diffusion-Based Synthetic Data Pretraining for Enhancing Activity Recognition

Evaluation details

CABiGRU pre-trained on diffusion-generated smartwatch IMU windows, fine-tuned on real data

Reported as of: 2026-10-05

90.6

Rows are ranked only within this task × dataset × metric. Protocols can differ; read each method's evaluation details before comparing. Latest reported row: 2026-10-05.

About the dataset

DEO: Drinking / Eating / Other smartwatch IMU (Samsung GW4 relabel)

Subjects
90
Signals
accel, gyro, magnetometer
Licence
Not stated

Sources on this board

Related leaderboards

Data and submissions

Download every leaderboard as JSON or CSV. Missing a published result? Email hello@chatppg.com with the paper link and the exact table it comes from.

Cite this

Cite the page URL and the date you accessed it. Please also cite the original publications listed above.

APA

ChatPPG. (2026). DEO balanced accuracy leaderboard: human activity recognition [Data set]. ChatPPG Research Hub. https://chatppg.com/research/leaderboards/har/deo-balanced-accuracy

BibTeX
@misc{chatppg_har_deo_balanced_accuracy,
  author       = {{ChatPPG}},
  title        = {{DEO balanced accuracy leaderboard: human activity recognition}},
  year         = {2026},
  month        = oct,
  howpublished = {\url{https://chatppg.com/research/leaderboards/har/deo-balanced-accuracy}},
  note         = {ChatPPG Research Hub}
}
CITATION.cff
cff-version: 1.2.0
message: "If you use this resource, please cite it as below."
type: dataset
title: "DEO balanced accuracy leaderboard: human activity recognition"
authors:
  - name: "ChatPPG"
url: "https://chatppg.com/research/leaderboards/har/deo-balanced-accuracy"
date-released: "2026-10-16"