MASPC_Transform: A Plant Point Cloud Segmentation Network Based on Multi-Head Attention Separation and Position Code
Open Access
Abstract
Plant point cloud segmentation is an important step in 3D plant phenotype research. Because the stems, leaves, flowers, and other organs of plants are often intertwined and small in size, this makes plant point cloud segmentation more challenging than other segmentation tasks. In this paper, we propose MASPC_Transform, a novel plant point cloud segmentation network base on multi-head attention separation and position code. The proposed MASPC_Transform establishes connections for similar point clouds scattered in different areas of the point cloud space through multiple attention heads. In order to avoid the aggregation of multiple attention heads, we propose a multi-head attention separation loss based on spatial similarity, so that the attention positions of different attention heads can be dispersed as much as possible. In order to reduce the impact of point cloud disorder and irregularity on feature extraction, we propose a new point cloud position coding method, and use the position coding network based on this method in the local and global feature extraction modules of MASPC_Transform. We evaluate our MASPC_Transform on the ROSE_X dataset. Compared with the state-of-the-art approaches, the proposed MASPC_Transform achieved better segmentation results.Keywords
Funding Information
- Department of Science and Technology of Jilin Province (20200201165JC)
This publication has 34 references indexed in Scilit:
- Deep Learning: Individual Maize Segmentation From Terrestrial Lidar Data Using Faster R-CNN and Regional Growth AlgorithmsFrontiers in Plant Science, 2018
- Retrieving 2-D Leaf Angle Distributions for Deciduous Trees From Terrestrial Laser Scanner DataIEEE Transactions on Geoscience and Remote Sensing, 2018
- Multi-view Convolutional Neural Networks for 3D Shape Recognition2015
- Segmenting tree crowns from terrestrial and mobile LiDAR data by exploring ecological theoriesISPRS Journal of Photogrammetry and Remote Sensing, 2015
- VoxNet: A 3D Convolutional Neural Network for real-time object recognition2015
- Tree Leaves Extraction in Natural Images: Comparative Study of Preprocessing Tools and Segmentation MethodsIEEE Transactions on Image Processing, 2015
- Adaptive clustering of airborne LiDAR data to segment individual tree crowns in managed pine forestsInternational Journal of Remote Sensing, 2010
- Plant Phenotype Demarcation Using Nontargeted LC-MS and GC-MS Metabolite ProfilingJournal of Agricultural and Food Chemistry, 2009
- Knowledge and heuristic-based modeling of laser-scanned treesACM Transactions on Graphics, 2007
- Regional cerebral perfusion abnormalities in attention deficit/hyperactivity disorderEuropean Archives of Psychiatry and Clinical Neuroscience, 2002