Performance Validation of Yolo Variants for Object Detection

Kaiyue Liu; Haitong Tang; Shuang He; Qin Yu; Yulong Xiong; Nizhuan Wang

doi:10.1145/3448748.3448786

Performance Validation of Yolo Variants for Object Detection

Kaiyue Liu, Haitong Tang, Shuang He, Qin Yu, Yulong Xiong, Nizhuan Wang

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

52 Citations (Scopus)

Abstract

Object detection is a core part of an intelligent surveillance system and a fundamental algorithm in the field of identity identification, which is of great practical importance. Since the YOLO series algorithms have good results in terms of accuracy and speed, YOLO and each subsequent version have been surpassing. Thus, in this paper, it carries out experiments on three versions of popular YOLO models such as yolov3, yolov4, and yolov5 (yolov5l, yolov5m, yolov5s, yolov5x). The performance of the three versions of YOLO model is analyzed and summarized by training and predicting the public VOC dataset. Results showed that the yolov4 model is higher than the yolov3 model in terms of mAP values, but slightly lower in terms of speed, while the yolov5 series model is better than the yolov3 and yolov4 models both in terms of mAP values and speed.

Original language	English
Title of host publication	Proceedings of the 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021
Publisher	Association for Computing Machinery, Inc
Pages	239-243
Number of pages	5
ISBN (Electronic)	9781450390002
DOIs	https://doi.org/10.1145/3448748.3448786
Publication status	Published - 21 Mar 2021
Externally published	Yes
Event	2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021 - Virtual, Online, China Duration: 22 Jan 2021 → 24 Jan 2021

Conference

Conference	2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021
Country/Territory	China
City	Virtual, Online
Period	22/01/21 → 24/01/21

Keywords

Deep Learning
Object Detection
PASCAL VOC Dataset
YOLO

ASJC Scopus subject areas

Health Informatics
Artificial Intelligence
Information Systems
Biomedical Engineering
Computational Theory and Mathematics

More information

10.1145/3448748.3448786

Cite this

@inproceedings{f7f691b0e55c4a6aa7c3bb178da0e925,

title = "Performance Validation of Yolo Variants for Object Detection",

abstract = "Object detection is a core part of an intelligent surveillance system and a fundamental algorithm in the field of identity identification, which is of great practical importance. Since the YOLO series algorithms have good results in terms of accuracy and speed, YOLO and each subsequent version have been surpassing. Thus, in this paper, it carries out experiments on three versions of popular YOLO models such as yolov3, yolov4, and yolov5 (yolov5l, yolov5m, yolov5s, yolov5x). The performance of the three versions of YOLO model is analyzed and summarized by training and predicting the public VOC dataset. Results showed that the yolov4 model is higher than the yolov3 model in terms of mAP values, but slightly lower in terms of speed, while the yolov5 series model is better than the yolov3 and yolov4 models both in terms of mAP values and speed.",

keywords = "Deep Learning, Object Detection, PASCAL VOC Dataset, YOLO",

author = "Kaiyue Liu and Haitong Tang and Shuang He and Qin Yu and Yulong Xiong and Nizhuan Wang",

note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021 ; Conference date: 22-01-2021 Through 24-01-2021",

year = "2021",

month = mar,

day = "21",

doi = "10.1145/3448748.3448786",

language = "English",

pages = "239--243",

booktitle = "Proceedings of the 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021",

publisher = "Association for Computing Machinery, Inc",

}

Liu, K, Tang, H, He, S, Yu, Q, Xiong, Y & Wang, N 2021, Performance Validation of Yolo Variants for Object Detection. in Proceedings of the 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021., 3448786, Association for Computing Machinery, Inc, pp. 239-243, 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021, Virtual, Online, China, 22/01/21. https://doi.org/10.1145/3448748.3448786

Performance Validation of Yolo Variants for Object Detection. / Liu, Kaiyue; Tang, Haitong; He, Shuang et al.
Proceedings of the 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021. Association for Computing Machinery, Inc, 2021. p. 239-243 3448786.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Performance Validation of Yolo Variants for Object Detection

AU - Liu, Kaiyue

AU - Tang, Haitong

AU - He, Shuang

AU - Yu, Qin

AU - Xiong, Yulong

AU - Wang, Nizhuan

PY - 2021/3/21

Y1 - 2021/3/21

N2 - Object detection is a core part of an intelligent surveillance system and a fundamental algorithm in the field of identity identification, which is of great practical importance. Since the YOLO series algorithms have good results in terms of accuracy and speed, YOLO and each subsequent version have been surpassing. Thus, in this paper, it carries out experiments on three versions of popular YOLO models such as yolov3, yolov4, and yolov5 (yolov5l, yolov5m, yolov5s, yolov5x). The performance of the three versions of YOLO model is analyzed and summarized by training and predicting the public VOC dataset. Results showed that the yolov4 model is higher than the yolov3 model in terms of mAP values, but slightly lower in terms of speed, while the yolov5 series model is better than the yolov3 and yolov4 models both in terms of mAP values and speed.

AB - Object detection is a core part of an intelligent surveillance system and a fundamental algorithm in the field of identity identification, which is of great practical importance. Since the YOLO series algorithms have good results in terms of accuracy and speed, YOLO and each subsequent version have been surpassing. Thus, in this paper, it carries out experiments on three versions of popular YOLO models such as yolov3, yolov4, and yolov5 (yolov5l, yolov5m, yolov5s, yolov5x). The performance of the three versions of YOLO model is analyzed and summarized by training and predicting the public VOC dataset. Results showed that the yolov4 model is higher than the yolov3 model in terms of mAP values, but slightly lower in terms of speed, while the yolov5 series model is better than the yolov3 and yolov4 models both in terms of mAP values and speed.

KW - Deep Learning

KW - Object Detection

KW - PASCAL VOC Dataset

KW - YOLO

UR - http://www.scopus.com/inward/record.url?scp=85103514824&partnerID=8YFLogxK

U2 - 10.1145/3448748.3448786

DO - 10.1145/3448748.3448786

M3 - Conference article published in proceeding or book

AN - SCOPUS:85103514824

SP - 239

EP - 243

BT - Proceedings of the 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021

PB - Association for Computing Machinery, Inc

T2 - 2021 International Conference on Bioinformatics and Intelligent Computing, BIC 2021

Y2 - 22 January 2021 through 24 January 2021

ER -

Performance Validation of Yolo Variants for Object Detection

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this