Title: “Deep Learning-Based Cognitive Assistance for Segmentation of Wires and Ethernet Devices in RGB Images”
Abstract: This article presents a deep learning-based approach for segmentation of wires and ethernet devices in RGB images. The proposed implementation is responsible for producing binary segmentation masks using the R50-FPN backbone network with the PointRend mask head. Two models were tested, and their predicted binary segmentation masks were used to segment wires and devices in the corresponding depth image to extract individual object point clouds. The results show that the proposed method can accurately segment wires and ethernet devices, making it suitable for use in augmented reality applications.
Reference: Wang, Y., Li, J., Li, Z., & Yu, H. (2021). Deep Learning-Based Cognitive Assistance for Segmentation of Wires and Ethernet Devices in RGB Images. IEEE Access, 9, 30415-30424. doi: 10.1109/ACCESS.2021.3061797
