Deep Point Cloud Normal Estimation via Triplet Learning
This addresses a fundamental issue in 3D vision and graphics for applications like CAD modeling, but it is incremental as it builds on existing deep learning approaches.
The paper tackles the problem of normal estimation on 3D point clouds, which often struggles with accuracy at sharp features and robustness to noise, by proposing a method that uses triplet learning for feature encoding and achieves better results on CAD-like shapes.
Normal estimation on 3D point clouds is a fundamental problem in 3D vision and graphics. Current methods often show limited accuracy in predicting normals at sharp features (e.g., edges and corners) and less robustness to noise. In this paper, we propose a novel normal estimation method for point clouds. It consists of two phases: (a) feature encoding which learns representations of local patches, and (b) normal estimation that takes the learned representation as input and regresses the normal vector. We are motivated that local patches on isotropic and anisotropic surfaces have similar or distinct normals, and that separable features or representations can be learned to facilitate normal estimation. To realise this, we first construct triplets of local patches on 3D point cloud data, and design a triplet network with a triplet loss for feature encoding. We then design a simple network with several MLPs and a loss function to regress the normal vector. Despite having a smaller network size compared to most other methods, experimental results show that our method preserves sharp features and achieves better normal estimation results on CAD-like shapes.