Title: Indoor object segmentation based on YOLACT++

Authors: Ying Sun; Zichen Zhao; Bo Tao; Xin Liu; Juntong Yun; Ying Liu

Addresses: Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education, Wuhan University of Science and Technology, Wuhan, Hubei, China; Research Centre for Biomimetic Robot and Intelligent Measurement and Control, Wuhan University of science and Technology, Wuhan, Hubei, China ' Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education, Wuhan University of Science and Technology, Wuhan, Hubei, China; Research Centre for Biomimetic Robot and Intelligent Measurement and Control, Wuhan University of science and Technology, Wuhan, Hubei, China ' Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education, Wuhan University of Science and Technology, Wuhan, Hubei, China; Precision Manufacturing Research Institute, Wuhan University of Science and Technology, Wuhan, Hubei, China ' Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education, Wuhan University of Science and Technology, Wuhan, Hubei, China; Precision Manufacturing Research Institute, Wuhan University of Science and Technology, Wuhan, Hubei, China ' Research Centre for Biomimetic Robot and Intelligent Measurement and Control, Wuhan University of Science and Technology, Wuhan, Hubei, China; Precision Manufacturing Research Institute, Wuhan University of Science and Technology, Wuhan, Hubei, China ' Research Centre for Biomimetic Robot and Intelligent Measurement and Control, Wuhan University of Science and Technology, Wuhan, Hubei, China; Precision Manufacturing Research Institute, Wuhan University of Science and Technology, Wuhan, Hubei, China

Abstract: Intelligent Manufacturing originates from the research of artificial intelligence, which can not only reduce operating costs, but also improve product quality, and has become the way to cultivate new advantages in forging international competition in manufacturing. With the transformation of traditional manufacturing to intelligent manufacturing, the course of intelligent manufacturing should also be different from the traditional teaching mode, but through the survey, it has been found that the course of intelligent manufacturing still has a single teaching mode, insufficient innovative guidance for students and insufficient combination of theory and practice. In order to solve the issues mentioned above, the Project-Analysis-Evaluation (PAE) structure is proposed in this paper, and it is combined with constructivist theory and analysed through the study of indoor object instance segmentation detection. The YOLACT++ algorithm is pre-trained on SUNRGBD dataset and applied to indoor environment detection in this paper.

Keywords: intelligent manufacturing; instance segmentation; YOLACT++; PAE-teaching under constructivism.

DOI: 10.1504/IJWMC.2023.130402

International Journal of Wireless and Mobile Computing, 2023 Vol.24 No.2, pp.127 - 133

Received: 11 Jun 2022
Received in revised form: 01 Jul 2022
Accepted: 13 Jul 2022
Published online: 19 Apr 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article