Next Article in Journal
A Photon Force and Flow for Dissipative Structuring: Application to Pigments, Plants and Ecosystems
Previous Article in Journal
Causality in Schwinger’s Picture of Quantum Mechanics
Article

Domain-Specific On-Device Object Detection Method

Department of Electronics and Communications Engineering, Kwangwoon University, Seoul 01897, Korea
*
Author to whom correspondence should be addressed.
Academic Editors: Andrea Prati, Carlos A. Iglesias, Luis Javier García Villalba and Vincent A. Cicirello
Entropy 2022, 24(1), 77; https://doi.org/10.3390/e24010077
Received: 26 November 2021 / Revised: 25 December 2021 / Accepted: 29 December 2021 / Published: 1 January 2022
(This article belongs to the Topic Machine and Deep Learning)
Object detection is a significant activity in computer vision, and various approaches have been proposed to detect varied objects using deep neural networks (DNNs). However, because DNNs are computation-intensive, it is difficult to apply them to resource-constrained devices. Here, we propose an on-device object detection method using domain-specific models. In the proposed method, we define object of interest (OOI) groups that contain objects with a high frequency of appearance in specific domains. Compared with the existing DNN model, the layers of the domain-specific models are shallower and narrower, reducing the number of trainable parameters; thus, speeding up the object detection. To ensure a lightweight network design, we combine various network structures to obtain the best-performing lightweight detection model. The experimental results reveal that the size of the proposed lightweight model is 21.7 MB, which is 91.35% and 36.98% smaller than those of YOLOv3-SPP and Tiny-YOLO, respectively. The f-measure achieved on the MS COCO 2017 dataset were 18.3%, 11.9% and 20.3% higher than those of YOLOv3-SPP, Tiny-YOLO and YOLO-Nano, respectively. The results demonstrated that the lightweight model achieved higher efficiency and better performance on non-GPU devices, such as mobile devices and embedded boards, than conventional models. View Full-Text
Keywords: object detection; domain-specific; on-device; lightweight network object detection; domain-specific; on-device; lightweight network
Show Figures

Figure 1

MDPI and ACS Style

Kang, S.; Hwang, J.; Chung, K. Domain-Specific On-Device Object Detection Method. Entropy 2022, 24, 77. https://doi.org/10.3390/e24010077

AMA Style

Kang S, Hwang J, Chung K. Domain-Specific On-Device Object Detection Method. Entropy. 2022; 24(1):77. https://doi.org/10.3390/e24010077

Chicago/Turabian Style

Kang, Seongju, Jaegi Hwang, and Kwangsue Chung. 2022. "Domain-Specific On-Device Object Detection Method" Entropy 24, no. 1: 77. https://doi.org/10.3390/e24010077

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop