MODNet: Multi-offset Point Cloud Denoising Network Customized for Multi-scale Patches
This addresses surface degradation issues in 3D point cloud processing, which is important for applications like computer vision and robotics, but it appears incremental as it builds on existing multi-scale patch methods.
The paper tackles the problem of surface degradation in point cloud denoising, such as remnant noise and loss of geometric details, by proposing MODNet, which uses multi-scale geometric perception to adaptively weight offsets, achieving state-of-the-art performance on synthetic and real-scanned datasets.
The intricacy of 3D surfaces often results cutting-edge point cloud denoising (PCD) models in surface degradation including remnant noise, wrongly-removed geometric details. Although using multi-scale patches to encode the geometry of a point has become the common wisdom in PCD, we find that simple aggregation of extracted multi-scale features can not adaptively utilize the appropriate scale information according to the geometric information around noisy points. It leads to surface degradation, especially for points close to edges and points on complex curved surfaces. We raise an intriguing question -- if employing multi-scale geometric perception information to guide the network to utilize multi-scale information, can eliminate the severe surface degradation problem? To answer it, we propose a Multi-offset Denoising Network (MODNet) customized for multi-scale patches. First, we extract the low-level feature of three scales patches by patch feature encoders. Second, a multi-scale perception module is designed to embed multi-scale geometric information for each scale feature and regress multi-scale weights to guide a multi-offset denoising displacement. Third, a multi-offset decoder regresses three scale offsets, which are guided by the multi-scale weights to predict the final displacement by weighting them adaptively. Experiments demonstrate that our method achieves new state-of-the-art performance on both synthetic and real-scanned datasets.