Asymmetrical Siamese Network for Point Clouds Normal Estimation
This addresses overfitting and cross-domain generalization for researchers in 3D computer vision, though it's incremental with a new dataset and architectural tweak.
The paper tackles overfitting in point cloud normal estimation by proposing an Asymmetric Siamese Network that enforces consistency between features from clean and noisy point clouds, and introduces a new multi-view dataset with varied shapes and noise levels. Experiments show the dataset challenges existing methods and their feature constraint mechanism improves performance and reduces overfitting.
In recent years, deep learning-based point cloud normal estimation has made great progress. However, existing methods mainly rely on the PCPNet dataset, leading to overfitting. In addition, the correlation between point clouds with different noise scales remains unexplored, resulting in poor performance in cross-domain scenarios. In this paper, we explore the consistency of intrinsic features learned from clean and noisy point clouds using an Asymmetric Siamese Network architecture. By applying reasonable constraints between features extracted from different branches, we enhance the quality of normal estimation. Moreover, we introduce a novel multi-view normal estimation dataset that includes a larger variety of shapes with different noise levels. Evaluation of existing methods on this new dataset reveals their inability to adapt to different types of shapes, indicating a degree of overfitting. Extensive experiments show that the proposed dataset poses significant challenges for point cloud normal estimation and that our feature constraint mechanism effectively improves upon existing methods and reduces overfitting in current architectures.