Optimized PointNet for 3D Object Classification

Li, Zhuangzhuang and Li, Wenmei and Liu, Haiyan and Wang, Yu and Gui, Guan (2019) Optimized PointNet for 3D Object Classification. In: Advanced Hybrid Information Processing. Third EAI International Conference, ADHIP 2019, Nanjing, China, September 21–22, 2019, Proceedings, Part I.

[thumbnail of 75078.pdf] PDF
75078.pdf

Download (389kB)

Abstract

Three-dimensional (3D) laser scanning technology is widely used to get the 3D geometric information of the surrounding environment, which leads to a huge increase interest of point cloud. The PointNet based on neural network can directly process point clouds, and it provides a unified frame to handle the task of object classification, part segmentation and semantic segmentation. It is indicated that the PointNet is efficient for target segmentation. However, the number of neural network layers and loss function are not good enough for target classification. In this paper, we optimize the original neural network by deepen the layers of neural network. Simulation result shows that the overall accuracy increases from 89.20% to 89.35%. Meanwhile, the combination of softmax loss with center loss function is adopt to enhance the robustness of classification, and the overall accuracy is up to 89.95%.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Date Deposited: 04 Mar 2026 19:15
Last Modified: 20 Nov 2026 11:20
URI: https://eprints.eai.eu/id/eprint/55734

Actions (login required)

View Item
View Item