MiniYOLO: A lightweight object detection algorithm that realizes the trade‐off between model size and detection accuracy

The object detection task is to locate and classify objects in an image. The current state‐of‐the‐art high‐accuracy object detection algorithms rely on complex networks and high computational cost. These algorithms have high requirements on the memory resource and computing capability of the deploye...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of intelligent systems Jg. 37; H. 12; S. 12135 - 12151
Hauptverfasser: Liu, Yi, Zhang, Changsheng, Wu, Wenjing, Zhang, Bin, Zhou, Fucai
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York John Wiley & Sons, Inc 01.12.2022
Schlagworte:
Accuracy
Algorithms
Computer networks
computer vision
depthwise separable convolutions
Electronic devices
Image classification
Intelligent systems
lightweight
Model accuracy
multiscale feature fusion
object detection
Object recognition
ISSN:0884-8173, 1098-111X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The object detection task is to locate and classify objects in an image. The current state‐of‐the‐art high‐accuracy object detection algorithms rely on complex networks and high computational cost. These algorithms have high requirements on the memory resource and computing capability of the deployed device, and are difficult to apply to mobile and embedded devices. Through the depthwise separable convolution and multiple efficient network structures, this paper designs a lightweight backbone network and two different multiscale feature fusion structures, and proposes a lightweight one‐stage object detection algorithm—MiniYOLO. With the model size of only 4.2 MB, MiniYOLO still maintains a high detection accuracy, realizing the trade‐off between the model size and detection accuracy. Experimental results on MS COCO 2017 data set show that compared to the state‐of‐the‐art PP‐YOLO‐tiny, MiniYOLO achieves higher mAP with the same model size. Compared with other lightweight object detection algorithms, MiniYOLO has certain advantages in detection accuracy or model size. The code associated with this paper can be downloaded from https://github.com/CaedmonLY/MiniYOLO/.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0884-8173
1098-111X
DOI:10.1002/int.23079