An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud Registration
Non-rigid transformation is based on rigid transformation by adding distortions to form a more complex but more consistent common scene. Many advanced non-rigid alignment models are implemented using supervised learning; however, the large number of labels required for the training process makes the...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/11/3525 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850129301988442112 |
|---|---|
| author | Munan Yuan Xiru Li Haibao Tan |
| author_facet | Munan Yuan Xiru Li Haibao Tan |
| author_sort | Munan Yuan |
| collection | DOAJ |
| description | Non-rigid transformation is based on rigid transformation by adding distortions to form a more complex but more consistent common scene. Many advanced non-rigid alignment models are implemented using supervised learning; however, the large number of labels required for the training process makes their application difficult. Here, an elastic fine-tuning dual recurrent computation for unsupervised non-rigid registration is proposed. At first, we transform a non-rigid transformation into a series of combinations of rigid transformations using an outer recurrent computational network. Then, the inner loop layer computes elastic-controlled rigid incremental transformations by controlling the threshold to obtain a finely coherent rigid transformation. Finally, we design and implement loss functions that constrain deformations and keep transformations as rigid as possible. Extensive experiments validate that the proposed method achieves state-of-the-art performance with 0.01219 earth mover’s distances (EMDs) and 0.0153 root mean square error (RMSE) in non-rigid and rigid scenes, respectively. |
| format | Article |
| id | doaj-art-145542a23cae4508a1708156cbfdcd70 |
| institution | OA Journals |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-145542a23cae4508a1708156cbfdcd702025-08-20T02:33:02ZengMDPI AGSensors1424-82202025-06-012511352510.3390/s25113525An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud RegistrationMunan Yuan0Xiru Li1Haibao Tan2Hefei Institute of Physical Science, Chinese Academy of Sciences, Hefei 230031, ChinaHefei Institute of Physical Science, Chinese Academy of Sciences, Hefei 230031, ChinaHefei Institute of Physical Science, Chinese Academy of Sciences, Hefei 230031, ChinaNon-rigid transformation is based on rigid transformation by adding distortions to form a more complex but more consistent common scene. Many advanced non-rigid alignment models are implemented using supervised learning; however, the large number of labels required for the training process makes their application difficult. Here, an elastic fine-tuning dual recurrent computation for unsupervised non-rigid registration is proposed. At first, we transform a non-rigid transformation into a series of combinations of rigid transformations using an outer recurrent computational network. Then, the inner loop layer computes elastic-controlled rigid incremental transformations by controlling the threshold to obtain a finely coherent rigid transformation. Finally, we design and implement loss functions that constrain deformations and keep transformations as rigid as possible. Extensive experiments validate that the proposed method achieves state-of-the-art performance with 0.01219 earth mover’s distances (EMDs) and 0.0153 root mean square error (RMSE) in non-rigid and rigid scenes, respectively.https://www.mdpi.com/1424-8220/25/11/3525non-rigid registration3D modelingunsupervisedelastic fine-tuningdual recurrent computation |
| spellingShingle | Munan Yuan Xiru Li Haibao Tan An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud Registration Sensors non-rigid registration 3D modeling unsupervised elastic fine-tuning dual recurrent computation |
| title | An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud Registration |
| title_full | An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud Registration |
| title_fullStr | An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud Registration |
| title_full_unstemmed | An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud Registration |
| title_short | An Elastic Fine-Tuning Dual Recurrent Framework for Non-Rigid Point Cloud Registration |
| title_sort | elastic fine tuning dual recurrent framework for non rigid point cloud registration |
| topic | non-rigid registration 3D modeling unsupervised elastic fine-tuning dual recurrent computation |
| url | https://www.mdpi.com/1424-8220/25/11/3525 |
| work_keys_str_mv | AT munanyuan anelasticfinetuningdualrecurrentframeworkfornonrigidpointcloudregistration AT xiruli anelasticfinetuningdualrecurrentframeworkfornonrigidpointcloudregistration AT haibaotan anelasticfinetuningdualrecurrentframeworkfornonrigidpointcloudregistration AT munanyuan elasticfinetuningdualrecurrentframeworkfornonrigidpointcloudregistration AT xiruli elasticfinetuningdualrecurrentframeworkfornonrigidpointcloudregistration AT haibaotan elasticfinetuningdualrecurrentframeworkfornonrigidpointcloudregistration |