Crowd-based people detection using deep learning
Human detection is a hot research field in computer vision. Its purpose is to detect humans in images or video sequences and provide precise location by using bounding boxes. One major challenge is to detect people in crowded environments. Specifically, this challenge is broken into two main diff...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Thesis-Master by Coursework |
Language: | English |
Published: |
Nanyang Technological University
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/155650 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-155650 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1556502023-07-04T17:36:15Z Crowd-based people detection using deep learning Chen, Lei Yap Kim Hui School of Electrical and Electronic Engineering EKHYap@ntu.edu.sg Engineering::Electrical and electronic engineering Human detection is a hot research field in computer vision. Its purpose is to detect humans in images or video sequences and provide precise location by using bounding boxes. One major challenge is to detect people in crowded environments. Specifically, this challenge is broken into two main difficulties: multi-scale problems and occlusion problems. Most object detection algorithms have low accuracy for small object detection and identification. When objects are occluded, the features will overlap, making it hard to determine a correct outcome and leading to false or missed detection. This project first reviewed an extensive list of literature related to object detection based on handcraft and deep learning methods. Then, two state of art neural networks were introduced (EfficientDet and YOLOv5), and through further analysis, I analyzed the components and the thesis and the subsequent source codes, deduced the complete network structure, and explained the specific implementation process of the critical parts. From this, I used the Crowd- Human data set to train the aforementioned networks, and got two object detection networks for people in crowded environments. Experimental results suggest that compared to Faster RCNN, EfficientDet yields detection speeds three times faster in crowded environments, whereas YOLOv5 provides the fastest detection speed while increasing AP by almost 18%. In addition, YOLOv5 solves the difficulty of detecting small-scale objects while maintaining high accuracy of objects in crowded environments. Overall, YOLOv5 can effectively detect people in crowded environments and prevent missed detection of hard-to-detect small-scale people. Its high detection accuracy and fast detection speed make an effective option for crowded environment scenarios. Keywords: Deep Learning, Object Detection, Human Detection. Master of Science (Signal Processing) 2022-03-10T04:20:40Z 2022-03-10T04:20:40Z 2021 Thesis-Master by Coursework Chen, L. (2021). Crowd-based people detection using deep learning. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/155650 https://hdl.handle.net/10356/155650 en ISM-DISS-02451 application/pdf Nanyang Technological University |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Electrical and electronic engineering |
spellingShingle |
Engineering::Electrical and electronic engineering Chen, Lei Crowd-based people detection using deep learning |
description |
Human detection is a hot research field in computer vision. Its purpose is to
detect humans in images or video sequences and provide precise location by
using bounding boxes. One major challenge is to detect people in crowded
environments. Specifically, this challenge is broken into two main difficulties:
multi-scale problems and occlusion problems. Most object detection algorithms
have low accuracy for small object detection and identification. When objects
are occluded, the features will overlap, making it hard to determine a correct
outcome and leading to false or missed detection.
This project first reviewed an extensive list of literature related to object detection
based on handcraft and deep learning methods. Then, two state of
art neural networks were introduced (EfficientDet and YOLOv5), and through
further analysis, I analyzed the components and the thesis and the subsequent
source codes, deduced the complete network structure, and explained the specific
implementation process of the critical parts. From this, I used the Crowd-
Human data set to train the aforementioned networks, and got two object detection
networks for people in crowded environments. Experimental results suggest
that compared to Faster RCNN, EfficientDet yields detection speeds three times
faster in crowded environments, whereas YOLOv5 provides the fastest detection
speed while increasing AP by almost 18%. In addition, YOLOv5 solves
the difficulty of detecting small-scale objects while maintaining high accuracy
of objects in crowded environments.
Overall, YOLOv5 can effectively detect people in crowded environments and
prevent missed detection of hard-to-detect small-scale people. Its high detection
accuracy and fast detection speed make an effective option for crowded environment
scenarios.
Keywords: Deep Learning, Object Detection, Human Detection. |
author2 |
Yap Kim Hui |
author_facet |
Yap Kim Hui Chen, Lei |
format |
Thesis-Master by Coursework |
author |
Chen, Lei |
author_sort |
Chen, Lei |
title |
Crowd-based people detection using deep learning |
title_short |
Crowd-based people detection using deep learning |
title_full |
Crowd-based people detection using deep learning |
title_fullStr |
Crowd-based people detection using deep learning |
title_full_unstemmed |
Crowd-based people detection using deep learning |
title_sort |
crowd-based people detection using deep learning |
publisher |
Nanyang Technological University |
publishDate |
2022 |
url |
https://hdl.handle.net/10356/155650 |
_version_ |
1772828925803626496 |