A Hierarchical Attention Fused Descriptor for 3D Point Matching
Motivated by recent successes on learning 3D feature representations, we present a Siamese network to generate representative 3D descriptors for 3D point matching in point cloud registration. Our system, dubbed HAF-Net, consists of feature extraction module, hierarchical feature reweighting and reca...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8733802/ |
| _version_ | 1818416083614302208 |
|---|---|
| author | Wenjun Shi Dongchen Zhu Liang Du Guanghui Zhang Jiamao Li Xiaolin Zhang |
| author_facet | Wenjun Shi Dongchen Zhu Liang Du Guanghui Zhang Jiamao Li Xiaolin Zhang |
| author_sort | Wenjun Shi |
| collection | DOAJ |
| description | Motivated by recent successes on learning 3D feature representations, we present a Siamese network to generate representative 3D descriptors for 3D point matching in point cloud registration. Our system, dubbed HAF-Net, consists of feature extraction module, hierarchical feature reweighting and recalibration module (HRR), as well as feature aggregation and compression module. The HRR module is proposed to adaptively integrate multi-level features through learning, acting as a hierarchical attention fusion mechanism. The learnable feature pooling technique VLAD is extended into our aggregation module, which is further utilized to extract principal components of features and compress them into a low dimensional feature vector. To train our model, we amass a large dataset for 3D point matching. The dataset is composed of matched and unmatched point block pairs, which are automatically searched from existing reconstruction datasets with known poses. The experiments demonstrate that the proposed HAF-Net not only outperforms other state-of-the-art approaches in 3D feature representation but also has a good generalization ability in various tasks and datasets. |
| first_indexed | 2024-12-14T11:45:15Z |
| format | Article |
| id | doaj.art-a0fbe7987b574c49a168bcd8dc1d3f91 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T11:45:15Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-a0fbe7987b574c49a168bcd8dc1d3f912022-12-21T23:02:38ZengIEEEIEEE Access2169-35362019-01-017774367744710.1109/ACCESS.2019.29218288733802A Hierarchical Attention Fused Descriptor for 3D Point MatchingWenjun Shi0https://orcid.org/0000-0003-2882-2981Dongchen Zhu1Liang Du2https://orcid.org/0000-0002-7952-5736Guanghui Zhang3https://orcid.org/0000-0003-0432-7329Jiamao Li4Xiaolin Zhang5Bionic Vision System Laboratory, State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, ChinaBionic Vision System Laboratory, State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, ChinaBionic Vision System Laboratory, State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, ChinaBionic Vision System Laboratory, State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, ChinaBionic Vision System Laboratory, State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, ChinaBionic Vision System Laboratory, State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, ChinaMotivated by recent successes on learning 3D feature representations, we present a Siamese network to generate representative 3D descriptors for 3D point matching in point cloud registration. Our system, dubbed HAF-Net, consists of feature extraction module, hierarchical feature reweighting and recalibration module (HRR), as well as feature aggregation and compression module. The HRR module is proposed to adaptively integrate multi-level features through learning, acting as a hierarchical attention fusion mechanism. The learnable feature pooling technique VLAD is extended into our aggregation module, which is further utilized to extract principal components of features and compress them into a low dimensional feature vector. To train our model, we amass a large dataset for 3D point matching. The dataset is composed of matched and unmatched point block pairs, which are automatically searched from existing reconstruction datasets with known poses. The experiments demonstrate that the proposed HAF-Net not only outperforms other state-of-the-art approaches in 3D feature representation but also has a good generalization ability in various tasks and datasets.https://ieeexplore.ieee.org/document/8733802/3D descriptorhierarchical attentiondata reweightingfeature recalibrationaggregationdata driven |
| spellingShingle | Wenjun Shi Dongchen Zhu Liang Du Guanghui Zhang Jiamao Li Xiaolin Zhang A Hierarchical Attention Fused Descriptor for 3D Point Matching IEEE Access 3D descriptor hierarchical attention data reweighting feature recalibration aggregation data driven |
| title | A Hierarchical Attention Fused Descriptor for 3D Point Matching |
| title_full | A Hierarchical Attention Fused Descriptor for 3D Point Matching |
| title_fullStr | A Hierarchical Attention Fused Descriptor for 3D Point Matching |
| title_full_unstemmed | A Hierarchical Attention Fused Descriptor for 3D Point Matching |
| title_short | A Hierarchical Attention Fused Descriptor for 3D Point Matching |
| title_sort | hierarchical attention fused descriptor for 3d point matching |
| topic | 3D descriptor hierarchical attention data reweighting feature recalibration aggregation data driven |
| url | https://ieeexplore.ieee.org/document/8733802/ |
| work_keys_str_mv | AT wenjunshi ahierarchicalattentionfuseddescriptorfor3dpointmatching AT dongchenzhu ahierarchicalattentionfuseddescriptorfor3dpointmatching AT liangdu ahierarchicalattentionfuseddescriptorfor3dpointmatching AT guanghuizhang ahierarchicalattentionfuseddescriptorfor3dpointmatching AT jiamaoli ahierarchicalattentionfuseddescriptorfor3dpointmatching AT xiaolinzhang ahierarchicalattentionfuseddescriptorfor3dpointmatching AT wenjunshi hierarchicalattentionfuseddescriptorfor3dpointmatching AT dongchenzhu hierarchicalattentionfuseddescriptorfor3dpointmatching AT liangdu hierarchicalattentionfuseddescriptorfor3dpointmatching AT guanghuizhang hierarchicalattentionfuseddescriptorfor3dpointmatching AT jiamaoli hierarchicalattentionfuseddescriptorfor3dpointmatching AT xiaolinzhang hierarchicalattentionfuseddescriptorfor3dpointmatching |