Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Sep 25, 2025

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

PFF-CB: Multiscale Occlusion Pedestrian Detection Method Based on PFF and CBAM.

Guiyi Yang¹, Zhengyou Wang², Shanna Zhuang¹

¹School of Information Science and Technology, Shijiazhuang Tiedao University, Shaoxing, China.

Computational Intelligence and Neuroscience

|May 2, 2022

Summary

This study introduces a novel Parallel Feature Fusion with CBAM (PFF-CB) network to improve occlusion pedestrian detection. The PFF-CB network effectively fuses multi-scale features, enhancing detection accuracy in crowded or severely occluded scenes.

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

Injectable NIR-II Photothermal Hydrogels for Image-Guided Adipose Remodeling and Antiobesity Therapy.

ACS nano·2026

Same author

Synthesis and Functionalization of π-Extended [7]Carbohelicene.

Chemistry, an Asian journal·2026

Same author

A FRET ratiometric fluorescent probe for detection of bisulfite in food: Insights into food quality and preservation.

Analytica chimica acta·2025

Same author

Staged Turnbull-Cutait pull-through anastomosis comparing with direct anastomosis plus prophylactic ileostomy in the treatment of low rectal cancer after internal sphincter resection (STAR-TAR): study protocol for a randomized controlled trial.

Trials·2025

Same author

Application of a fluorescent probe exhibiting a large stokes shift after the precise detection of hydrazine in plants, foods and living cells.

Journal of hazardous materials·2025

Same author

Coumarin-aurone based fluorescence probes for cysteine sensitive in-situ identification in living cells.

Colloids and surfaces. B, Biointerfaces·2024

Area of Science:

Computer Vision
Artificial Intelligence
Machine Learning

Background:

Occlusion pedestrian detection is a critical challenge in computer vision.
Existing methods using pedestrian parts or human body relationships struggle with severe occlusion and crowded scenes.
These traditional approaches fail to effectively utilize limited visible information.

Purpose of the Study:

To address the limitations of current methods in occlusion pedestrian detection.
To propose a novel network architecture for improved detection accuracy under occlusion.
To enhance the fusion of multi-scale features for better representation of occluded pedestrians.

Main Methods:

A new multiscale feature attention fusion network, Parallel Feature Fusion with CBAM (PFF-CB), is proposed.

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

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

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

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Sep 25, 2025

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

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

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

Published on: December 15, 2023

The PFF-CB module integrates feature information from different scales effectively.

It employs a Convolutional Block Attention Module (CBAM) for spatial and channel feature enhancement and a FPN-based parallel fusion module for key feature strengthening.

Main Results:

The PFF-CB module demonstrated strong performance on two common occlusion pedestrian datasets.
The network successfully integrated multi-scale features, improving detection in challenging scenarios.
Attention mechanisms within CBAM enhanced the focus on relevant spatial and channel information.

Conclusions:

The proposed PFF-CB network offers a significant advancement in occlusion pedestrian detection.
The method effectively handles varying scales and limited visible parts of pedestrians.
This approach shows promise for real-world applications requiring robust pedestrian detection in complex environments.