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 Concept Videos

Bone Structure01:55

Bone Structure

Within the skeletal system, the structure of a bone, or osseous tissue, can be exemplified in a long bone, like the femur, where there are two types of osseous tissue: cortical and cancellous.
Bone Remodeling01:40

Bone Remodeling

Bone remodeling is a continuous and balanced process of bone resorption by osteoclasts and bone formation by osteoblasts. In adults, it helps maintain bone mass and calcium homeostasis. While mechanical stress can stimulate turnover as part of the normal maintenance and reparative process, several hormones also regulate bone remodeling.
Classification of Bones01:18

Classification of Bones

The bones of the human skeletal system are of varied shapes, sizes, and functions. They can be classified based on their shape and function into four major classes: long bones, short bones, flat bones, and irregular bones. Some classifications include a fifth type, the sesamoid bones, as a separate class, whereas others categorize them under short bones.
Long and Short Bones
The appendicular skeleton, particularly the upper and lower limbs, is primarily made of long and short bones. The long...
Bone Remodeling and Repair01:31

Bone Remodeling and Repair

Osteoclasts are cells responsible for bone resorption and remodeling. They originate from hematopoietic progenitor cells present in the bone marrow. Numerous progenitor cells fuse to form multinucleated cells, each with 10-20 nuclei. A single osteoclast has a diameter of 150 to 200 µM. These cells have ruffled borders that break down the underlying bone tissue and release minerals such as calcium into the blood in bone resorption. Osteoclasts cling to bones with their ruffled edges during bone...

You might also read

Related Articles

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

Sort by
Same author

Tenacibaculum xiamenense sp. nov., an algicidal bacterium isolated from coastal seawater.

International journal of systematic and evolutionary microbiology·2013
Same author

The anchoring protein SAP97 influences the trafficking and localisation of multiple membrane channels.

Biochimica et biophysica acta·2013
Same author

Aegilops tauschii draft genome sequence reveals a gene repertoire for wheat adaptation.

Nature·2013
Same author

Draft genome of the wheat A-genome progenitor Triticum urartu.

Nature·2013
Same author

Citreoviridin enhances tumor necrosis factor-α-induced adhesion of human umbilical vein endothelial cells.

Toxicology and industrial health·2013
Same author

Th17/Treg imbalance induced by increased incidence of atherosclerosis in patients with systemic lupus erythematosus (SLE).

Clinical rheumatology·2013

Related Experiment Video

Updated: Jul 12, 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

Latent space improved masked reconstruction model for human skeleton-based action recognition.

Enqing Chen1, Xueting Wang1, Xin Guo1

  • 1School of Electrical and Information Engineering, Zhengzhou University, Zhengzhou, China.

Frontiers in Neurorobotics
|March 28, 2025
PubMed
Summary

This study enhances human skeleton-based action recognition by improving encoder feature learning. New models, SkeletonMVAE and SkeletonMVQVAE, leverage variational autoencoder latent spaces for better classification and generalization, especially with limited data.

Keywords:
human skeleton-based action recognitionmasked reconstruction modelself-supervised learningvariational autoencodervector quantized variational autoencoder

More Related Videos

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

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

Related Experiment Videos

Last Updated: Jul 12, 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

Automated Joint Space Detection Improves Bone Segmentation Accuracy
06:45

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Published on: November 28, 2025

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

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Human skeleton-based action recognition is crucial in computer vision.
  • Masked autoencoders (MAE) excel at data reconstruction but struggle with feature extraction for classification.
  • Existing autoencoder structures limit encoder performance in visual classification tasks.

Purpose of the Study:

  • To enhance the feature extraction capabilities of encoders in action recognition tasks.
  • To improve classification performance and generalization ability using advanced autoencoder techniques.
  • To address the limitations of standard autoencoders in learning discriminative features for skeleton-based action recognition.

Main Methods:

  • Proposed SkeletonMVAE model, leveraging Variational Autoencoder (VAE) latent space to constrain features as distributions.
  • Developed SkeletonMVQVAE model, utilizing Vector Quantized Variational Autoencoder (VQVAE) latent space for discrete feature representation.
  • Integrated VAE and VQVAE latent spaces to enhance encoder's ability to learn deeper data structures.

Main Results:

  • Both SkeletonMVAE and SkeletonMVQVAE significantly improved encoder classification accuracy on NTU-60 and NTU-120 datasets.
  • SkeletonMVAE demonstrated stronger classification ability.
  • SkeletonMVQVAE showed superior generalization capabilities, particularly in low-data regimes.

Conclusions:

  • The proposed VAE and VQVAE-based approaches effectively enhance feature representation for skeleton-based action recognition.
  • These methods improve both classification accuracy and generalization, offering a robust solution for computer vision tasks.
  • SkeletonMVAE and SkeletonMVQVAE provide valuable alternatives for action recognition, especially when dealing with limited labeled data.