Next Article in Journal
Three-Dimensional Vulnerability Assessment of Peanut (Arachis hypogaea) Based on Comprehensive Drought Index and Vulnerability Surface: A Case Study of Shandong Province, China
Previous Article in Journal
Sea Surface Salinity Inversion Model for Changjiang Estuary and Adjoining Sea Area with SMAP and MODIS Data Based on Machine Learning and Preliminary Application
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 14
Issue 21
10.3390/rs14215356
Review Report
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

PU-WGCN: Point Cloud Upsampling Using Weighted Graph Convolutional Networks

Remote Sens. 2022, 14(21), 5356; https://doi.org/10.3390/rs14215356
by Fan Gu 1, Changlun Zhang 1,*, Hengyou Wang 1, Qiang He 1 and Lianzhi Huo 2
Reviewer 1:
Tim Morris
Reviewer 2:
Cihan Altuntas
Remote Sens. 2022, 14(21), 5356; https://doi.org/10.3390/rs14215356
Submission received: 6 August 2022 / Revised: 15 October 2022 / Accepted: 23 October 2022 / Published: 26 October 2022

Round 1

Reviewer 1 Report


Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript is related to point cloud densification. The raw point cloud has sparse and unevenly distributed set of points. The current densification algorithms have a problem on a hole filling and boundary while upsampling the point cloud. They act all the nodes which have similar properties and characteristics belong to similar features. The manuscript proposed a point cloud upsampling network named PU-WGCN to overcome the crossing the border of the nodes and hole filling. The results are given with the Figures, and numerical results of Chamfer distance (CD), Hausdroff distance (HD) and Point to surface (P2F) as the measure of upsampling network accuracy.

The Figures and quantitative comparison of the results are shown the strength of the manuscript. But it is not clear how to get distances of CD,HD and P2F. How did you decide point-to-point or point-to-surface conjunction to estimate the distances. It has any information about this. It should make the manuscript better. Give a brief detail about it. (As an option, I offer point cloud comparison with iterative closest point (ICP) algorithm to the numerical results.)

I offer minor revisions. Short information about the mentioned points (given above 2nd) makes the manuscript better. It is a good study. It can be ACCEPTED for publication.

1. The real image can be added to the figures about examples (for example Fig.11 and the other group studies)

2. (This propose is optional, you can ignore it.) You can give numerical results about the success of your and other methods by comparing the unsampling point cloud with ground-truth data. The comparison can be made with ICP registration residuals. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper addresses a very interesting topic, the research methodology is well explained and results are clear.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Back to TopTop