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

You might also read

Related Articles

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

Sort by
Same author

YOLOv9-Based Detection of Diseases in Poplar Trees Using Histogram Equalization and Computer Vision.

Sensors (Basel, Switzerland)·2026
Same author

PatientEase-Domain-Aware RAG for Rehabilitation Instruction Simplification.

Bioengineering (Basel, Switzerland)·2025
Same author

A Generative Expert-Narrated Simplification Model for Enhancing Health Literacy Among the Older Population.

Bioengineering (Basel, Switzerland)·2025
Same author

From Anatomy to Genomics Using a Multi-Task Deep Learning Approach for Comprehensive Glioma Profiling.

Bioengineering (Basel, Switzerland)·2025
Same author

From Pixels to Precision-A Dual-Stream Deep Network for Pathological Nuclei Segmentation.

Bioengineering (Basel, Switzerland)·2025
Same author

Bridging the Gap Between Accuracy and Efficiency in AI-Based Breast Cancer Diagnosis from Histopathological Data.

Cancers·2025

Related Experiment Video

Updated: Aug 15, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

601

Improved Face Detection Method via Learning Small Faces on Hard Images Based on a Deep Learning Approach.

Dilnoza Mamieva1, Akmalbek Bobomirzaevich Abdusalomov1, Mukhriddin Mukhiddinov1,2

  • 1Department of Computer Engineering, Gachon University, Sujeong-gu, Seongnam-si 461-701, Gyeonggi-do, Republic of Korea.

Sensors (Basel, Switzerland)
|January 8, 2023
PubMed
Summary

This study introduces RetinaNet baseline, a deep learning face detector, to improve accuracy in challenging conditions like occlusion and blur. The model achieves high performance on benchmarks, enhancing face detection capabilities.

Keywords:
deep learningface detectionregion offering networkretina net

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.3K

Related Experiment Videos

Last Updated: Aug 15, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

601
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.9K
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.3K

Area of Science:

  • Computer Vision
  • Deep Learning
  • Artificial Intelligence

Background:

  • Traditional face detection methods struggle with uncontrolled environments and occluded faces.
  • Deep learning has significantly advanced computer vision, including face detection accuracy.
  • Challenges remain in detecting small, blurred, or partially occluded faces.

Purpose of the Study:

  • To propose RetinaNet baseline, a single-stage face detector, to address limitations in current face detection systems.
  • To enhance detection speed and accuracy for faces in uncontrolled conditions.
  • To improve the performance of face detection software.

Main Methods:

  • Implemented a single-stage face detector based on the RetinaNet architecture.
  • Introduced network improvements to boost detection speed and accuracy.
  • Trained and evaluated the model using PyTorch on WIDER FACE and FDDB datasets.

Main Results:

  • Achieved an Average Precision (AP) of 41.0 at 11.8 FPS (single-scale) and 44.2 AP (multi-scale) on the WIDER FACE benchmark.
  • Attained 95.6% accuracy in detecting faces using the PyTorch framework.
  • Demonstrated superior performance compared to existing one-stage detectors.

Conclusions:

  • The proposed RetinaNet baseline effectively handles challenging face detection scenarios.
  • Network improvements lead to significant gains in both speed and accuracy.
  • The model shows competitive and superior results on standard face detection benchmarks.