Journal article 434 views 219 downloads
C2SPoint: A classification-to-saliency network for point cloud saliency detection
Computers and Graphics, Volume: 115, Pages: 274 - 284
Swansea University Author:
Gary Tam
-
PDF | Accepted Manuscript
For the purpose of Open Access the author has applied a CC BY copyright licence to any Author Accepted Manuscript version arising from this submission.
Download (5.2MB)
DOI (Published version): 10.1016/j.cag.2023.07.003
Abstract
Point cloud saliency detection is an important technique that support downstream tasks in 3D graphics and vision, like 3D model simplification, compression, reconstruction and viewpoint selection. Existing approaches often rely on hand-crafted features and are only applicable to specific datasets. I...
Published in: | Computers and Graphics |
---|---|
ISSN: | 0097-8493 0097-8493 |
Published: |
Elsevier BV
2023
|
Online Access: |
Check full text
|
URI: | https://cronfa.swan.ac.uk/Record/cronfa63869 |
Abstract: |
Point cloud saliency detection is an important technique that support downstream tasks in 3D graphics and vision, like 3D model simplification, compression, reconstruction and viewpoint selection. Existing approaches often rely on hand-crafted features and are only applicable to specific datasets. In this paper, we propose a novel weakly supervised classification network, called C2SPoint, which directly performs saliency detection on the point clouds. Unlike previous methods that require per-point saliency annotations, C2SPoint only requires category labels of the point clouds during training. The network consists of two branches: a Classification branch and a Saliency branch. The former branch is composed of two Adaptive Set Abstraction layers for feature extraction and a Saliency Transform layer for learning saliency knowledge from the classification network. The latter branch introduces a multi-scale point-cluster similarity matrix for propagating the cluster saliency to each point within it, resulting in the prediction of point-level saliency. Experimental results demonstrate the effectiveness of our method in point cloud saliency detection, with improvements of 2% in both AUC and NSS compared to state-of-the-art methods. |
---|---|
Keywords: |
Point cloud saliency detection, Weakly-supervised learning, Learning features, PointNet++ |
College: |
Faculty of Science and Engineering |
Funders: |
This research was partially supported by Zhejiang Province Natural Science Foundation No. LY21F020013, LY22F020013, the National Natural Science Foundation of China No. 62172366. Gary Tam is supported by the Royal Society grant IEC/NSFC/211159. For the purpose of Open Access the author has applied a CC BY copyright licence to any Author Accepted Manuscript version arising from this submission. |
Start Page: |
274 |
End Page: |
284 |