Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

3D Point Capsule Networks

About

In this paper, we propose 3D point-capsule networks, an auto-encoder designed to process sparse 3D point clouds while preserving spatial arrangements of the input data. 3D capsule networks arise as a direct consequence of our novel unified 3D auto-encoder formulation. Their dynamic routing scheme and the peculiar 2D latent space deployed by our approach bring in improvements for several common point cloud-related tasks, such as object classification, object reconstruction and part segmentation as substantiated by our extensive evaluations. Moreover, it enables new applications such as part interpolation and replacement.

Yongheng Zhao, Tolga Birdal, Haowen Deng, Federico Tombari• 2018

Related benchmarks

TaskDatasetResultRank
3D Object ClassificationModelNet40 (test)
Accuracy93.2
302
3D Point Cloud ClassificationModelNet40 (test)
OA88.9
297
3D Shape ClassificationModelNet40 (test)
Accuracy89.3
227
Object ClassificationModelNet40 (test)--
180
3D Object Part SegmentationShapeNet Part (test)--
114
ClassificationModelNet40 (test)
Accuracy88.9
99
Few-shot 3D Object Classification (5-way)ModelNet40 (test)
10-shot Accuracy42.3
57
Few-shot 3D Object Classification (10-way)ModelNet40 (test)
Accuracy (10-shot)38
46
Few-shot object classificationModelNet40 (test)
Accuracy (5-way, 10-shot)42.3
29
Descriptor matching3DMatch Rotated--
18
Showing 10 of 16 rows

Other info

Code

Follow for update