Boundary–Inner Disentanglement Enhanced Learning for Point Cloud Semantic Segmentation

Round 1

Reviewer 1 Report

This article proposed a novel lightweight BIDEL framework that can improve the semantic segmentation accuracy of boundary points. The idea is simple and interesting. However, the paper suffers the following limits, authors should carefully revise this manuscript before its publication:

1. In Introduction, if the authors could introduce some methods considering boundary information, that would be better. As I know there are many neural networks considering the boundary. Therefore, what is different between their approaches and yours?

2. Besides, in Abstract and Introduction, please highlight the motivation of this work.  For example, what are the disadvantages of relying on additional models?

3. In Sec. 3, please carefully edit the formula in this session. 

4. In Sec. 4, RandLA-Net and KPConv are selected as baselines. Since from Table 2, JSENet is obviously better than proposed. If the authors could add some more methods considering boundary, that would be better. Besides, please explain the advantages of the proposed algorithm compared with JSENet.

5.  In Sec. 4.3, except for visual comparison, are there any quantitative results with previous metrics on Semantic3D dataset? 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Section 3 must be improved.

-       At the beginning of the section add a summary of your methodology

-       You must properly introduce the equation, list in detail the variables contained in it with a concise description of the meaning. To make them more readable show them in a bulleted list. In this way the reader will be able to understand the contribution of each variable.

-       200) mIoU Introduce adequately this type of metric, furthermore, do not use acronyms until you have presented the full definition, I will not repeat this advice again, it also applies to the other occurrences

-       201) OA, mACC, mIoU Introduce adequately this type of metrics

-       244) Figure 2 must be improved: add a label to all subplots and add a description for each subplot in the caption

-       The section relating to the methodologies based on Machine Learning must be enriched. You must summarize the essential characteristics of the methods you have used and justify your choices. Try to summarize what are the strengths and weaknesses of the methods, in this way you can make the reader understand why you have chosen these methodologies.

-       I could not find a detailed description of the evaluation metrics you have adopted. How will you measure your model's performance? This section is essential in order to demonstrate the effectiveness of your methodology. Furthermore, only by adopting adequate metrics will it be possible to compare your results with those obtained by other researchers.

Section 4 must be improved.

-       In this section you present the results of applying your methodology

-       You have detailed the data used to test your methodology

-       You should do the same in describing your methodology

-       In the case of Machine Learning applications, data quality is crucial for the success of the analyses.

-       A detailed description of the sample size on which the model was trained is missing. How much data did you have available? How did you split them between training and testing?

-       How did you set the parameters of the algorithm?

-       Did you perform a hyperparameter optimization?

-       A description of the hardware and software used for data processing is completely missing. Describe in detail the hardware used:  Extract this data from the datasheet of the hardware manufacturer. To make reading the specifications of the hardware more immediate, you can insert them in a table, listing the instruments used and the specific characteristics for each. Also, you should describe in detail the software platform you used. Also describe the machine learning-based libraries you used.

-       A detailed discussion of the results obtained is missing. Try to summarize what was obtained and try to extract useful information from the work carried out. Also add bibliographic references to support your conclusions, to give more weight to your statements.

Section 5 must be improved.

-       Paragraphs are missing where the possible practical applications of the results of this study are reported. What these results can serve the people, it is necessary to insert possible uses of this study that justify their publication.

-       They also lack the possible future goals of this work. Do the authors plan to continue their research on this topic?

 

60) Do not use abbreviation such as i.e. ,I will not repeat this advice again, it also applies to the other occurrences.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I appreciate the author’s efforts and their response to the previous version’s comments that let my concerns dispel. If authors could do further English editing and beautify the experimental presentation, that manuscript could be better.

Reviewer 2 Report

The authors addressed the reviewer's comments with attention and modified the paper with the suggestions provided. The new version of the paper has improved both in the presentation and in the contents