A Lightweight Structure Based on Feature Fusion for Point Cloud Analysis
Point cloud analysis is challenging due to the irregularity and sparsity, making it difficult to capture the underlying geometric characteristics. This paper proposes a lightweight structure that effectively aggregates the local patterns and the spatial layout of them extracted from the irregular an...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9521484/ |
| _version_ | 1818358070018834432 |
|---|---|
| author | Qiang Zheng Jian Sun |
| author_facet | Qiang Zheng Jian Sun |
| author_sort | Qiang Zheng |
| collection | DOAJ |
| description | Point cloud analysis is challenging due to the irregularity and sparsity, making it difficult to capture the underlying geometric characteristics. This paper proposes a lightweight structure that effectively aggregates the local patterns and the spatial layout of them extracted from the irregular and sparse point cloud. Unlike previous works that seek sophisticated feature extraction methods, the key to this structure is simultaneously exploring the local features and their distribution features in a concise manner. Specifically, the two features, which correspond to two different scales, are extracted independently and then fused. In this way, an abstract shape-level feature that contains much shape awareness and robustness is obtained. Moreover, since the structure exhibits rapid convergence with the conventional exponential decay learning schedule, we upgrade it by introducing the snapshot ensemble and creatively design a more flexible and effective cyclic annealing learning schedule. We evaluate the structure on challenging benchmarks, and experiment results prove that our model achieves on-par performance with previous state-of-the-art (SOTA) methods, although with the simple shared multi-layer perceptrons (MLPs) as feature extractors. |
| first_indexed | 2024-12-13T20:23:09Z |
| format | Article |
| id | doaj.art-c934474af7b64a5da4985e4ad153c788 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-13T20:23:09Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-c934474af7b64a5da4985e4ad153c7882022-12-21T23:32:39ZengIEEEIEEE Access2169-35362021-01-01911865111866110.1109/ACCESS.2021.31072859521484A Lightweight Structure Based on Feature Fusion for Point Cloud AnalysisQiang Zheng0https://orcid.org/0000-0001-9480-5827Jian Sun1https://orcid.org/0000-0003-3104-7904State Key Laboratory for Strength and Vibration of Mechanical Structures, School of Aerospace Engineering, Xi’an Jiaotong University, Xi’an, ChinaState Key Laboratory for Strength and Vibration of Mechanical Structures, School of Aerospace Engineering, Xi’an Jiaotong University, Xi’an, ChinaPoint cloud analysis is challenging due to the irregularity and sparsity, making it difficult to capture the underlying geometric characteristics. This paper proposes a lightweight structure that effectively aggregates the local patterns and the spatial layout of them extracted from the irregular and sparse point cloud. Unlike previous works that seek sophisticated feature extraction methods, the key to this structure is simultaneously exploring the local features and their distribution features in a concise manner. Specifically, the two features, which correspond to two different scales, are extracted independently and then fused. In this way, an abstract shape-level feature that contains much shape awareness and robustness is obtained. Moreover, since the structure exhibits rapid convergence with the conventional exponential decay learning schedule, we upgrade it by introducing the snapshot ensemble and creatively design a more flexible and effective cyclic annealing learning schedule. We evaluate the structure on challenging benchmarks, and experiment results prove that our model achieves on-par performance with previous state-of-the-art (SOTA) methods, although with the simple shared multi-layer perceptrons (MLPs) as feature extractors.https://ieeexplore.ieee.org/document/9521484/Point clouddeep learninglightweight networkclassification |
| spellingShingle | Qiang Zheng Jian Sun A Lightweight Structure Based on Feature Fusion for Point Cloud Analysis IEEE Access Point cloud deep learning lightweight network classification |
| title | A Lightweight Structure Based on Feature Fusion for Point Cloud Analysis |
| title_full | A Lightweight Structure Based on Feature Fusion for Point Cloud Analysis |
| title_fullStr | A Lightweight Structure Based on Feature Fusion for Point Cloud Analysis |
| title_full_unstemmed | A Lightweight Structure Based on Feature Fusion for Point Cloud Analysis |
| title_short | A Lightweight Structure Based on Feature Fusion for Point Cloud Analysis |
| title_sort | lightweight structure based on feature fusion for point cloud analysis |
| topic | Point cloud deep learning lightweight network classification |
| url | https://ieeexplore.ieee.org/document/9521484/ |
| work_keys_str_mv | AT qiangzheng alightweightstructurebasedonfeaturefusionforpointcloudanalysis AT jiansun alightweightstructurebasedonfeaturefusionforpointcloudanalysis AT qiangzheng lightweightstructurebasedonfeaturefusionforpointcloudanalysis AT jiansun lightweightstructurebasedonfeaturefusionforpointcloudanalysis |