Spherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representation
Coarse-to-fine 3D instance segmentation methods show weak performances compared to recent Grouping-based, Kernel-based and Transformer-based methods. We argue that this is due to two limitations: 1) Instance size over-estimation by axis-aligned bounding box(AABB) 2) False negative error accumulation...
| Main Authors: | , , , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
IEEE
2024
|
| _version_ | 1826316825929973760 |
|---|---|
| author | Shin, S Zhou, K Vankadari, M Markham, A Trigoni, N |
| author_facet | Shin, S Zhou, K Vankadari, M Markham, A Trigoni, N |
| author_sort | Shin, S |
| collection | OXFORD |
| description | Coarse-to-fine 3D instance segmentation methods show weak performances compared to recent Grouping-based, Kernel-based and Transformer-based methods. We argue that this is due to two limitations: 1) Instance size over-estimation by axis-aligned bounding box(AABB) 2) False negative error accumulation from inaccurate box to the re-finement phase. In this work, we introduce Spherical Mask, a novel coarse-to-fine approach based on spherical repre-sentation, overcoming those two limitations with several benefits. Specifically, our coarse detection estimates each in-stance with a 3D polygon using a center and radial distance predictions, which avoids excessive size estimation of AABB. To cut the error propagation in the existing coarse-to-fine approaches, we virtually migrate points based on the polygon, allowing all foreground points, including false negatives, to be refined. During inference, the proposal and point migration modules run in parallel and are assembled to form binary masks of instances. We also introduce two margin-based losses for the point migration to enforce corrections for the false positives/negatives and cohesion of foreground points, significantly improving the performance. Experimental results from three datasets, such as ScanNetV2, S3DIS, and STPLS3D, show that our proposed method outperforms existing works, demonstrating the effectiveness of the new in-stance representation with spherical coordinates. The code is available at: https://github.com/yunshin/SphericalMask |
| first_indexed | 2024-03-07T08:29:44Z |
| format | Conference item |
| id | oxford-uuid:cac82b34-5c5f-4458-bbd0-e7a304141bbc |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-02-19T04:30:36Z |
| publishDate | 2024 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:cac82b34-5c5f-4458-bbd0-e7a304141bbc2024-12-16T11:01:37ZSpherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representationConference itemhttp://purl.org/coar/resource_type/c_5794uuid:cac82b34-5c5f-4458-bbd0-e7a304141bbcEnglishSymplectic ElementsIEEE2024Shin, SZhou, KVankadari, MMarkham, ATrigoni, NCoarse-to-fine 3D instance segmentation methods show weak performances compared to recent Grouping-based, Kernel-based and Transformer-based methods. We argue that this is due to two limitations: 1) Instance size over-estimation by axis-aligned bounding box(AABB) 2) False negative error accumulation from inaccurate box to the re-finement phase. In this work, we introduce Spherical Mask, a novel coarse-to-fine approach based on spherical repre-sentation, overcoming those two limitations with several benefits. Specifically, our coarse detection estimates each in-stance with a 3D polygon using a center and radial distance predictions, which avoids excessive size estimation of AABB. To cut the error propagation in the existing coarse-to-fine approaches, we virtually migrate points based on the polygon, allowing all foreground points, including false negatives, to be refined. During inference, the proposal and point migration modules run in parallel and are assembled to form binary masks of instances. We also introduce two margin-based losses for the point migration to enforce corrections for the false positives/negatives and cohesion of foreground points, significantly improving the performance. Experimental results from three datasets, such as ScanNetV2, S3DIS, and STPLS3D, show that our proposed method outperforms existing works, demonstrating the effectiveness of the new in-stance representation with spherical coordinates. The code is available at: https://github.com/yunshin/SphericalMask |
| spellingShingle | Shin, S Zhou, K Vankadari, M Markham, A Trigoni, N Spherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representation |
| title | Spherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representation |
| title_full | Spherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representation |
| title_fullStr | Spherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representation |
| title_full_unstemmed | Spherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representation |
| title_short | Spherical mask: coarse-to-fine 3D point cloud instance segmentation with spherical representation |
| title_sort | spherical mask coarse to fine 3d point cloud instance segmentation with spherical representation |
| work_keys_str_mv | AT shins sphericalmaskcoarsetofine3dpointcloudinstancesegmentationwithsphericalrepresentation AT zhouk sphericalmaskcoarsetofine3dpointcloudinstancesegmentationwithsphericalrepresentation AT vankadarim sphericalmaskcoarsetofine3dpointcloudinstancesegmentationwithsphericalrepresentation AT markhama sphericalmaskcoarsetofine3dpointcloudinstancesegmentationwithsphericalrepresentation AT trigonin sphericalmaskcoarsetofine3dpointcloudinstancesegmentationwithsphericalrepresentation |