SoftPool++: An Encoder–Decoder Network for Point Cloud Completion

We propose a novel convolutional operator for the task of point cloud completion. One striking characteristic of our approach is that, conversely to related work it does not require any max-pooling or voxelization operation. Instead, the proposed operator used to learn the point cloud embedding in t...

Full description

Saved in:
Bibliographic Details
Published in:International journal of computer vision Vol. 130; no. 5; pp. 1145 - 1164
Main Authors: Wang, Yida, Tan, David Joseph, Navab, Nassir, Tombari, Federico
Format: Journal Article
Language:English
Published: New York Springer US 01.05.2022
Springer
Springer Nature B.V
Subjects:
Artificial Intelligence
Cloud computing
Coders
Computer Imaging
Computer Science
Feature extraction
Image Processing and Computer Vision
Neural networks
Operators (mathematics)
Pattern Recognition
Pattern Recognition and Graphics
Permutations
Special Issue on 3D Computer Vision
Vision
Skip-connection
SoftPool
Point cloud
Completion
ISSN:0920-5691, 1573-1405
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We propose a novel convolutional operator for the task of point cloud completion. One striking characteristic of our approach is that, conversely to related work it does not require any max-pooling or voxelization operation. Instead, the proposed operator used to learn the point cloud embedding in the encoder extracts permutation-invariant features from the point cloud via a soft-pooling of feature activations, which are able to preserve fine-grained geometric details. These features are then passed on to a decoder architecture. Due to the compression in the encoder, a typical limitation of this type of architectures is that they tend to lose parts of the input shape structure. We propose to overcome this limitation by using skip connections specifically devised for point clouds, where links between corresponding layers in the encoder and the decoder are established. As part of these connections, we introduce a transformation matrix that projects the features from the encoder to the decoder and vice-versa. The quantitative and qualitative results on the task of object completion from partial scans on the ShapeNet dataset show that incorporating our approach achieves state-of-the-art performance in shape completion both at low and high resolutions.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0920-5691
1573-1405
DOI:10.1007/s11263-022-01588-7