Exploring State Space Models in LiDAR Point Cloud Segmentation
Mamba has achieved significant success in various fields due to its ability to efficiently model long-range dependencies with linear complexity. However, its application in LiDAR point cloud processing is still in its early stages, facing challenges such as unordered and irregular data structures. I...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2025-07-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1015/2025/isprs-archives-XLVIII-G-2025-1015-2025.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839609330744164352 |
|---|---|
| author | D. Lu L. Xu R. Wang J. Li |
| author_facet | D. Lu L. Xu R. Wang J. Li |
| author_sort | D. Lu |
| collection | DOAJ |
| description | Mamba has achieved significant success in various fields due to its ability to efficiently model long-range dependencies with linear complexity. However, its application in LiDAR point cloud processing is still in its early stages, facing challenges such as unordered and irregular data structures. In this study, we investigated the performance of two existing Mamba-based algorithms, PointMamba and PointCloudMamba, on the aerial DALES LiDAR dataset for point cloud segmentation, and further explored the critical role of token serialization in influencing Mamba’s performance. To evaluate serialization quality, we proposed two novel indicators—Neighbor Preservation Ratio (NPR) and Sequence Jump Distance (SJD)—which quantify the ability of serialization methods to preserve spatial topology and geometric relationships. Our findings confirm the great potential of Mamba in LiDAR point cloud processing, and demonstrate that serialization significantly impacts Mamba’s performance, with better preservation of spatial and geometric relationships leading to higher segmentation accuracy. These results provide meaningful insights into improving Mamba’s performance in LiDAR point cloud processing and guiding the development of advanced serialization methods. |
| format | Article |
| id | doaj-art-34db315987494978b901b1b243d68687 |
| institution | Matheson Library |
| issn | 1682-1750 2194-9034 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj-art-34db315987494978b901b1b243d686872025-07-30T09:22:10ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342025-07-01XLVIII-G-20251015102110.5194/isprs-archives-XLVIII-G-2025-1015-2025Exploring State Space Models in LiDAR Point Cloud SegmentationD. Lu0L. Xu1R. Wang2J. Li3Department of Systems Design Engineering, University of Waterloo, Waterloo, Ontario N2L 3G1, CanadaDepartmemt of Geomatics Engineering, University of Calgary, Calgary, Alberta T2N 1N4, CanadaSchool of Architecture & Urban Planning, Shenzhen University, Shenzhen, GD 518060, ChinaDepartment of Geography and Environmental Management, University of Waterloo, Waterloo, ON N2L 3G1, CanadaMamba has achieved significant success in various fields due to its ability to efficiently model long-range dependencies with linear complexity. However, its application in LiDAR point cloud processing is still in its early stages, facing challenges such as unordered and irregular data structures. In this study, we investigated the performance of two existing Mamba-based algorithms, PointMamba and PointCloudMamba, on the aerial DALES LiDAR dataset for point cloud segmentation, and further explored the critical role of token serialization in influencing Mamba’s performance. To evaluate serialization quality, we proposed two novel indicators—Neighbor Preservation Ratio (NPR) and Sequence Jump Distance (SJD)—which quantify the ability of serialization methods to preserve spatial topology and geometric relationships. Our findings confirm the great potential of Mamba in LiDAR point cloud processing, and demonstrate that serialization significantly impacts Mamba’s performance, with better preservation of spatial and geometric relationships leading to higher segmentation accuracy. These results provide meaningful insights into improving Mamba’s performance in LiDAR point cloud processing and guiding the development of advanced serialization methods.https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1015/2025/isprs-archives-XLVIII-G-2025-1015-2025.pdf |
| spellingShingle | D. Lu L. Xu R. Wang J. Li Exploring State Space Models in LiDAR Point Cloud Segmentation The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | Exploring State Space Models in LiDAR Point Cloud Segmentation |
| title_full | Exploring State Space Models in LiDAR Point Cloud Segmentation |
| title_fullStr | Exploring State Space Models in LiDAR Point Cloud Segmentation |
| title_full_unstemmed | Exploring State Space Models in LiDAR Point Cloud Segmentation |
| title_short | Exploring State Space Models in LiDAR Point Cloud Segmentation |
| title_sort | exploring state space models in lidar point cloud segmentation |
| url | https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1015/2025/isprs-archives-XLVIII-G-2025-1015-2025.pdf |
| work_keys_str_mv | AT dlu exploringstatespacemodelsinlidarpointcloudsegmentation AT lxu exploringstatespacemodelsinlidarpointcloudsegmentation AT rwang exploringstatespacemodelsinlidarpointcloudsegmentation AT jli exploringstatespacemodelsinlidarpointcloudsegmentation |