Related Experiment Video
Updated: Sep 25, 2025
08:32
Tracking Rats in Operant Conditioning Chambers Using a Versatile Homemade Video Camera and DeepLabCut
Published on: June 15, 2020
12.7K
VTag: a semi-supervised pipeline for tracking pig activity with a single top-view camera
Chun-Peng J Chen1, Gota Morota2,3, Kiho Lee4
1Department of Animal Science, University of California, Davis, CA 95616, USA.
Journal of Animal Science
|April 29, 2022
Summary
A new semi-supervised pipeline, VTag, enables efficient pig activity monitoring using computer vision (CV) with minimal human labeling. This tool simplifies tracking and analysis for precision livestock farming.
Area of Science:
- Agricultural Engineering
- Computer Vision
- Animal Science
Background:
- Precision livestock farming is crucial for meeting global meat demand.
- Computer vision (CV) automates pig activity monitoring but faces challenges with variable illumination and extensive labeling requirements for supervised learning.
Purpose of the Study:
- To develop a semi-supervised pipeline (VTag) for efficient, long-term pig activity tracking.
- To reduce the need for pre-labeled data in CV systems for swine management.
Main Methods:
- Developed VTag, a semi-supervised pipeline requiring minimal human supervision.
- Utilized a single top-view RGB camera for pig tracking.
- Created a user-friendly software tool with a graphical interface.
Main Results:
- Achieved an average tracking error of 17.99 cm across datasets.
- Enabled estimation of pig movement distance per unit time for activity studies.
- Generated heat maps indicating spatial usage patterns for farming management.
Conclusions:
- VTag significantly reduces laborious dataset preparation for CV systems.
- The rapid deployment capability of VTag facilitates pig behavior monitoring in precision livestock farming.
