In this paper, we propose a normal estimation method for unstructured point cloud. We observe that geometric estimators commonly focus more on feature preservation but are hard to tune parameters and sensitive to noise, while learning-based approaches pursue an overall normal estimation accuracy but cannot well handle challenging regions such as surface edges. This paper presents a novel normal estimation method, under the co-support of geometric estimator and deep learning. To lowering the learning difficulty, we first propose to compute a suboptimal initial normal at each point by searching for a best fitting patch. Based on the computed normal field, we design a normal-based height map network (NH-Net) to fine-tune the suboptimal normals. Qualitative and quantitative evaluations demonstrate the clear improvements of our results over both traditional methods and learning-based methods, in terms of estimation accuracy and feature recovery.
|Title of host publication||The Proceedings of the Conference on Computer Vision and Pattern Recognition 2020 (CVPR 2020)|
|Number of pages||10|
|Publication status||Published - 14 Jun 2020|
|Event||The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2020 - Online|
Duration: 14 Jun 2020 → 19 Jun 2020
|Name||Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition|
|Publisher||IEEE Computer Society|
|Public Lecture||The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2020|
|Period||14/06/20 → 19/06/20|
Bibliographical noteFunding Information:
This work was supported by the National Natural Science Foundation of China (No. 61502137), the Hong Kong Research Grants Council (No. PolyU 152035/17E), the HKIBS Research Seed Fund 2019/20 (No. 190-009), and the Research Seed Fund (No. 102367) of Lingnan University, Hong Kong.