Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive Detection
High-fidelity mesh reconstruction from point clouds has long been a fundamental research topic in computer vision and computer graphics. Traditional methods require dense triangle meshes to achieve high fidelity, but excessively dense triangles may lead to unnecessary storage and computational burde...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/12/3155 |
| _version_ | 1797592791626285056 |
|---|---|
| author | Qi Liu Shibiao Xu Jun Xiao Ying Wang |
| author_facet | Qi Liu Shibiao Xu Jun Xiao Ying Wang |
| author_sort | Qi Liu |
| collection | DOAJ |
| description | High-fidelity mesh reconstruction from point clouds has long been a fundamental research topic in computer vision and computer graphics. Traditional methods require dense triangle meshes to achieve high fidelity, but excessively dense triangles may lead to unnecessary storage and computational burdens, while also struggling to capture clear, sharp, and continuous edges. This paper argues that the key to high-fidelity reconstruction lies in preserving sharp features. Therefore, we introduce a novel sharp-feature-preserving reconstruction framework based on primitive detection. It includes an improved deep-learning-based primitive detection module and two novel mesh splitting and selection modules that we propose. Our framework can accurately and reasonably segment primitive patches, fit meshes in each patch, and split overlapping meshes at the triangle level to ensure true sharpness while obtaining lightweight mesh models. Quantitative and visual experimental results demonstrate that our framework outperforms both the state-of-the-art learning-based primitive detection methods and traditional reconstruction methods. Moreover, our designed modules are plug-and-play, which not only apply to learning-based primitive detectors but also can be combined with other point cloud processing tasks such as edge extraction or random sample consensus (RANSAC) to achieve high-fidelity results. |
| first_indexed | 2024-03-11T01:58:14Z |
| format | Article |
| id | doaj.art-a46568a994164ad1957649c7595a97f8 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T01:58:14Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-a46568a994164ad1957649c7595a97f82023-11-18T12:27:15ZengMDPI AGRemote Sensing2072-42922023-06-011512315510.3390/rs15123155Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive DetectionQi Liu0Shibiao Xu1Jun Xiao2Ying Wang3School of Artificial Intelligence, University of Chinese Academy of Sciences, No. 19 Yuquan Road, Shijingshan District, Beijing 100049, ChinaSchool of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, ChinaSchool of Artificial Intelligence, University of Chinese Academy of Sciences, No. 19 Yuquan Road, Shijingshan District, Beijing 100049, ChinaSchool of Artificial Intelligence, University of Chinese Academy of Sciences, No. 19 Yuquan Road, Shijingshan District, Beijing 100049, ChinaHigh-fidelity mesh reconstruction from point clouds has long been a fundamental research topic in computer vision and computer graphics. Traditional methods require dense triangle meshes to achieve high fidelity, but excessively dense triangles may lead to unnecessary storage and computational burdens, while also struggling to capture clear, sharp, and continuous edges. This paper argues that the key to high-fidelity reconstruction lies in preserving sharp features. Therefore, we introduce a novel sharp-feature-preserving reconstruction framework based on primitive detection. It includes an improved deep-learning-based primitive detection module and two novel mesh splitting and selection modules that we propose. Our framework can accurately and reasonably segment primitive patches, fit meshes in each patch, and split overlapping meshes at the triangle level to ensure true sharpness while obtaining lightweight mesh models. Quantitative and visual experimental results demonstrate that our framework outperforms both the state-of-the-art learning-based primitive detection methods and traditional reconstruction methods. Moreover, our designed modules are plug-and-play, which not only apply to learning-based primitive detectors but also can be combined with other point cloud processing tasks such as edge extraction or random sample consensus (RANSAC) to achieve high-fidelity results.https://www.mdpi.com/2072-4292/15/12/3155mesh reconstructionpoint cloudssharp featureprimitive detection |
| spellingShingle | Qi Liu Shibiao Xu Jun Xiao Ying Wang Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive Detection Remote Sensing mesh reconstruction point clouds sharp feature primitive detection |
| title | Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive Detection |
| title_full | Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive Detection |
| title_fullStr | Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive Detection |
| title_full_unstemmed | Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive Detection |
| title_short | Sharp Feature-Preserving 3D Mesh Reconstruction from Point Clouds Based on Primitive Detection |
| title_sort | sharp feature preserving 3d mesh reconstruction from point clouds based on primitive detection |
| topic | mesh reconstruction point clouds sharp feature primitive detection |
| url | https://www.mdpi.com/2072-4292/15/12/3155 |
| work_keys_str_mv | AT qiliu sharpfeaturepreserving3dmeshreconstructionfrompointcloudsbasedonprimitivedetection AT shibiaoxu sharpfeaturepreserving3dmeshreconstructionfrompointcloudsbasedonprimitivedetection AT junxiao sharpfeaturepreserving3dmeshreconstructionfrompointcloudsbasedonprimitivedetection AT yingwang sharpfeaturepreserving3dmeshreconstructionfrompointcloudsbasedonprimitivedetection |