We design an end-to-end deep neural network architecture for LiDAR point cloud and 2D image point-wise feature fusion, which is suitable for directly consuming unordered point cloud. To the best of our knowledge, this is the first approach to use multimodal fusion network for aerial point cloud 3D segmentation which well respects the permutation invariance of point cloud. PMNet has an advantage over 2D based models that it can incorporate multi-view 3D scanned data if available.
Citation:
Poliyapram, V.; Wang, W.; Nakamura, R. A Point-Wise LiDAR and Image Multimodal Fusion Network (PMNet) for Aerial Point Cloud 3D Semantic Segmentation. Remote Sens. 2019, 11, 2961. https://www.mdpi.com/2072-4292/11/24/2961