Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 16, 2026

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats
06:52

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats

Published on: April 3, 2026

Multi-stream activated adaptive graph network based on incomplete skeletons for thermal adaptive behavior

Wenjun Duan1,2, Cunqian Wang1,2, Chaoqun Zheng3,4

  • 1School of Computer Science and Artificial Intelligence, Shandong Jianzhu University, Jinan, 250101, China.

Scientific Reports
|July 14, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Carbon Nanotube-Based Multi-Functional "Yin"-"Yang" Film: Dual Applications in Solar Water Evaporation and Blue Energy Harvesting.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

21 novel pathogenic variants identified in a cohort of 77 Chinese families with osteogenesis imperfecta.

Frontiers in genetics·2026
Same author

CO<sub>2</sub> adsorbent for sustainable development: synthesis route and performance study of slag-derived layered double oxides.

Journal of environmental management·2026
Same author

Stepwise Modified Transcrestal Sinus Floor Elevation in Sites With 3-5-mm Residual Bone Height: A Retrospective Study.

Clinical implant dentistry and related research·2026
Same author

Dual-anion NiSSe electrocatalyst enabling sulfur-tolerant sulfion oxidation for energy-efficient hydrogen production.

Journal of colloid and interface science·2026
Same author

Effects of plasma potassium on myocardial function: a POTCAST substudy.

Journal of cardiovascular imaging·2026

This study introduces a novel network (MTAGCN) for recognizing thermal adaptive behaviors from incomplete skeleton data. The model demonstrates superior accuracy and robustness, even with occluded or missing joint information, advancing thermal comfort perception.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Human-Computer Interaction

Background:

  • Video-based thermal comfort prediction research is expanding.
  • Existing methods struggle with skeleton occlusion and incomplete joint detection in real-world surveillance.
  • This limits accurate thermal adaptive behavior recognition.

Purpose of the Study:

  • To propose a robust method for thermal adaptive behavior recognition using incomplete skeleton sequences.
  • To enhance the performance of models under challenging occlusion conditions.
  • To develop a non-intrusive visual sensing framework for dynamic thermal comfort perception.

Main Methods:

  • A multi-stream activated adaptive graph network (MTAGCN) was developed.
  • A Transformer-enhanced adaptive graph convolution module was designed to capture spatial-temporal correlations.
Keywords:
Incomplete skeletonNon-intrusive monitoringThermal adaptation behaviorThermal comfort prediction

More Related Videos

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

Related Experiment Videos

Last Updated: Jul 16, 2026

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats
06:52

Deep-Learning Based Multi-Joint Synchronous Tracking for Objective Quantification of Hindlimb Locomotor Kinematics in Rats

Published on: April 3, 2026

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

  • Synthetic occlusion datasets were created to test model robustness.

    • Main Results:

    • MTAGCN significantly outperformed baseline methods on both original and occlusion-contaminated datasets.
    • The model achieved superior recognition accuracy and robustness against skeleton incompleteness.
    • Demonstrated effectiveness in handling frame, partial body, and random joint occlusions.

    Conclusions:

    • MTAGCN offers a robust solution for thermal adaptive behavior recognition from incomplete skeleton data.
    • The proposed framework is low-cost and non-intrusive for dynamic thermal comfort perception.
    • This research supports intelligent indoor environmental regulation by integrating behavioral recognition and thermal comfort prediction.