Research Interests
The point cloud data acquired by LiDAR is an important data source for scene perception. With the continuous development of high-precision mapping, geographic information multimodal data fusion and smart cities, there is an increasing demand for semantic segmentation of indoor and outdoor 3D point cloud data. Semantic segmentation based on 3D point clouds is the core technology to realize the optimization and upgrading of environmental perception. Meanwhile, semantic segmentation can assist in the incremental update of semantic maps. Environmental point cloud maps with semantic information can be well suited to the visual localization of autonomous vehicles.
My current research interests are mainly focused on point cloud analysis and processing techniques, aiming to explore high precision, low memory consumption methods for semantic segmentation of 3D point clouds.
Education
- MSc, Information and Communication Engineering, Jimei University, September 2019 – June 2022
- BSc, Software Engineering, Hubei University of Arts and Science, Sept 2015 – June 2019
Publication
- Jing Du, Guorong Cai, Zongyue Wang, et al. 2021. ResDLPS-Net: Joint residual-dense optimization for large-scale point cloud semantic segmentation. ISPRS Journal of Photogrammetry and Remote Sensing 182, 37–51. doi:10.1016/j.isprsjprs.2021.09.024.
- Jing Du, Zuning Jiang, Shangfeng Huang, et al. 2021. Point cloud semantic segmentation network based on multi-scale feature fusion. Sensors 21, 1625. doi:10.3390/s21051625.
- Jing Du, Guorong Cai, Zongyue Wang, et al. 2021. Convertible sparse convolution for point cloud instance segmentation. IGARSS. doi: 10.1109/IGARSS47720.2021.9553064.