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

Fixed Action Patterns01:06

Fixed Action Patterns

16.0K
A fixed action pattern (FAP) is a specific, hard-wired sequence of behaviors that occurs in response to an external stimulus, called a sign stimulus. The behavior is “fixed” because it is essentially unchangeable—proceeding similarly across individuals of a species every time it occurs.
16.0K

You might also read

Related Articles

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

Sort by
Same author

Advances in Video Emotion Recognition: Challenges and Trends.

Sensors (Basel, Switzerland)·2025
Same author

Genomic analyses identify agents regulating somatotroph and lactotroph functions.

Functional & integrative genomics·2016
Same author

Enhanced Procoagulant Activity on Blood Cells after Acute Ischemic Stroke.

Translational stroke research·2016
Same author

Hepatic FTO expression is increased in NASH and its silencing attenuates palmitic acid-induced lipotoxicity.

Biochemical and biophysical research communications·2016
Same author

Integrating microRNA and mRNA expression profiles of acute promyelocytic leukemia cells to explore the occurrence mechanisms of differentiation syndrome.

Oncotarget·2016
Same author

Fumaric Acid Production from Alkali-Pretreated Corncob by Fed-Batch Simultaneous Saccharification and Fermentation Combined with Separated Hydrolysis and Fermentation at High Solids Loading.

Applied biochemistry and biotechnology·2016

Related Experiment Video

Updated: May 2, 2026

Corticospinal Excitability Modulation During Action Observation
12:33

Corticospinal Excitability Modulation During Action Observation

Published on: December 31, 2013

8.9K

FineTea: A Novel Fine-Grained Action Recognition Video Dataset for Tea Ceremony Actions.

Changwei Ouyang1, Yun Yi1, Hanli Wang2

  • 1School of Mathematics and Computer Science, Gannan Normal University, Ganzhou 341000, China.

Journal of Imaging
|September 27, 2024
PubMed
Summary

Researchers developed a new deep learning method for fine-grained action recognition in videos. Their TSM-ConvNeXt model significantly improves accuracy on complex tasks like tea ceremonies.

Keywords:
ConvNeXtfine-grained action recognitiontea ceremony actionstemporal shift module

More Related Videos

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.6K

Related Experiment Videos

Last Updated: May 2, 2026

Corticospinal Excitability Modulation During Action Observation
12:33

Corticospinal Excitability Modulation During Action Observation

Published on: December 31, 2013

8.9K
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.6K

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Deep learning excels at video action recognition but struggles with fine-grained analysis in real-world scenarios.
  • Existing methods face limitations when applied to complex, subtle actions like those in a tea ceremony.

Purpose of the Study:

  • To advance fine-grained video action recognition capabilities.
  • To introduce a new dataset and a novel deep learning model tailored for intricate action classification.

Main Methods:

  • A hierarchical fine-grained action classification approach was used.
  • A new dataset, FineTea, comprising 2745 video clips of tea ceremony actions, was created.
  • A novel method, TSM-ConvNeXt, integrating Temporal Shift Module (TSM) with ConvNeXt, was proposed.

Main Results:

  • The TSM-ConvNeXt model achieved a 7.31% performance improvement over a ResNet50 baseline.
  • The proposed method demonstrated superior performance compared to state-of-the-art approaches on the FineTea and Diving48 datasets.
  • The FineTea dataset, featuring 9 basic and 31 fine-grained action subclasses, is now publicly available.

Conclusions:

  • The TSM-ConvNeXt model offers a significant advancement in fine-grained video action recognition.
  • The FineTea dataset provides a valuable resource for future research in detailed action understanding.
  • The proposed approach sets a new benchmark for recognizing complex, subtle actions in video data.