PBP-Net: Point Projection and Back-Projection Network for 3D Point Cloud Segmentation
This addresses the problem of dense prediction in 3D vision for applications like object part and indoor scene segmentation, but it is incremental as it builds on existing 2D CNN approaches.
The paper tackles 3D point cloud segmentation by proposing PBP-Net, a simple architecture that projects 3D points to 2D planes, uses 2D CNNs for feature extraction, and back-projects features to 3D, achieving comparable performance to state-of-the-art methods on ShapeNet-Part and S3DIS datasets.
Following considerable development in 3D scanning technologies, many studies have recently been proposed with various approaches for 3D vision tasks, including some methods that utilize 2D convolutional neural networks (CNNs). However, even though 2D CNNs have achieved high performance in many 2D vision tasks, existing works have not effectively applied them onto 3D vision tasks. In particular, segmentation has not been well studied because of the difficulty of dense prediction for each point, which requires rich feature representation. In this paper, we propose a simple and efficient architecture named point projection and back-projection network (PBP-Net), which leverages 2D CNNs for the 3D point cloud segmentation. 3 modules are introduced, each of which projects 3D point cloud onto 2D planes, extracts features using a 2D CNN backbone, and back-projects features onto the original 3D point cloud. To demonstrate effective 3D feature extraction using 2D CNN, we perform various experiments including comparison to recent methods. We analyze the proposed modules through ablation studies and perform experiments on object part segmentation (ShapeNet-Part dataset) and indoor scene semantic segmentation (S3DIS dataset). The experimental results show that proposed PBP-Net achieves comparable performance to existing state-of-the-art methods.