Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face Datasets
Owing to imperfect scans, occlusions, low reflectance of the scanned surface, and packet loss, there may be several incomplete regions in the 3D point cloud dataset. These missing regions can degrade the performance of recognition, classification, segmentation, or upsampling methods in point cloud d...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/3/723 |
| _version_ | 1797484901100945408 |
|---|---|
| author | Kutub Uddin Tae Hyun Jeong Byung Tae Oh |
| author_facet | Kutub Uddin Tae Hyun Jeong Byung Tae Oh |
| author_sort | Kutub Uddin |
| collection | DOAJ |
| description | Owing to imperfect scans, occlusions, low reflectance of the scanned surface, and packet loss, there may be several incomplete regions in the 3D point cloud dataset. These missing regions can degrade the performance of recognition, classification, segmentation, or upsampling methods in point cloud datasets. In this study, we propose a new approach to estimate the incomplete regions of 3D point cloud human face datasets using the masking method. First, we perform some preprocessing on the input point cloud, such as rotation in the left and right angles. Then, we project the preprocessed point cloud onto a 2D surface and generate masks. Finally, we interpolate the 2D projection and the mask to produce the estimated point cloud. We also designed a deep learning model to restore the estimated point cloud to improve its quality. We use chamfer distance (CD) and hausdorff distance (HD) to evaluate the proposed method on our own human face and large-scale facial model (LSFM) datasets. The proposed method achieves an average CD and HD results of 1.30 and 21.46 for our own and 1.35 and 9.08 for the LSFM datasets, respectively. The proposed method shows better results than the existing methods. |
| first_indexed | 2024-03-09T23:12:04Z |
| format | Article |
| id | doaj.art-bac9ef7a77fe4b7c8c2714f85306605c |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T23:12:04Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-bac9ef7a77fe4b7c8c2714f85306605c2023-11-23T17:44:15ZengMDPI AGSensors1424-82202022-01-0122372310.3390/s22030723Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face DatasetsKutub Uddin0Tae Hyun Jeong1Byung Tae Oh2School of Electronics and Information Engineering, Korea Aerospace University, Goyang 10540, KoreaSchool of Electronics and Information Engineering, Korea Aerospace University, Goyang 10540, KoreaSchool of Electronics and Information Engineering, Korea Aerospace University, Goyang 10540, KoreaOwing to imperfect scans, occlusions, low reflectance of the scanned surface, and packet loss, there may be several incomplete regions in the 3D point cloud dataset. These missing regions can degrade the performance of recognition, classification, segmentation, or upsampling methods in point cloud datasets. In this study, we propose a new approach to estimate the incomplete regions of 3D point cloud human face datasets using the masking method. First, we perform some preprocessing on the input point cloud, such as rotation in the left and right angles. Then, we project the preprocessed point cloud onto a 2D surface and generate masks. Finally, we interpolate the 2D projection and the mask to produce the estimated point cloud. We also designed a deep learning model to restore the estimated point cloud to improve its quality. We use chamfer distance (CD) and hausdorff distance (HD) to evaluate the proposed method on our own human face and large-scale facial model (LSFM) datasets. The proposed method achieves an average CD and HD results of 1.30 and 21.46 for our own and 1.35 and 9.08 for the LSFM datasets, respectively. The proposed method shows better results than the existing methods.https://www.mdpi.com/1424-8220/22/3/7233D point cloudincomplete regiondeep learningestimationand restoration |
| spellingShingle | Kutub Uddin Tae Hyun Jeong Byung Tae Oh Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face Datasets Sensors 3D point cloud incomplete region deep learning estimation and restoration |
| title | Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face Datasets |
| title_full | Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face Datasets |
| title_fullStr | Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face Datasets |
| title_full_unstemmed | Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face Datasets |
| title_short | Incomplete Region Estimation and Restoration of 3D Point Cloud Human Face Datasets |
| title_sort | incomplete region estimation and restoration of 3d point cloud human face datasets |
| topic | 3D point cloud incomplete region deep learning estimation and restoration |
| url | https://www.mdpi.com/1424-8220/22/3/723 |
| work_keys_str_mv | AT kutubuddin incompleteregionestimationandrestorationof3dpointcloudhumanfacedatasets AT taehyunjeong incompleteregionestimationandrestorationof3dpointcloudhumanfacedatasets AT byungtaeoh incompleteregionestimationandrestorationof3dpointcloudhumanfacedatasets |