Next Article in Journal
HDL-ODPRs: A Hybrid Deep Learning Technique Based Optimal Duplication Detection for Pull-Requests in Open-Source Repositories
Next Article in Special Issue
An Accurate, Efficient, and Stable Perspective-n-Point Algorithm in 3D Space
Previous Article in Journal
Back Analysis of Surrounding Rock Parameters of Large-Span Arch Cover Station Based on GP-DE Algorithm
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 12
Issue 24
10.3390/app122412593
Review Report
applsci-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

LUMDE: Light-Weight Unsupervised Monocular Depth Estimation via Knowledge Distillation

Appl. Sci. 2022, 12(24), 12593; https://doi.org/10.3390/app122412593
by Wenze Hu 1, Xue Dong 1,*, Ning Liu 2 and Yuanfeng Chen 2
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Appl. Sci. 2022, 12(24), 12593; https://doi.org/10.3390/app122412593
Submission received: 12 November 2022 / Revised: 6 December 2022 / Accepted: 7 December 2022 / Published: 8 December 2022
(This article belongs to the Special Issue 3D Scene Understanding and Object Recognition)

Round 1

Reviewer 1 Report

The paper presents unsupervised monocular depth estimation using knowledge distillation approach. The background of the research problem and the methodology is comprehensively presented. Overall, the paper fulfils the merits for publication. I have a few suggestions before that:

1. The authors emphasize the contribution by of their work inspired by its possible application in robots, autonomous vehicles. However, the results are shown only for indoor scenes. I would recommend to add results similar to Fig. 2 for outdoor scenes as well such as traffic/roads.  

2. The Related Work section need to be rewritten. The citation of studies are inappropriately placed. The style to cite a study is correct in Introduction section. The same should be followed in the Related Work section. 

3. Few sentences are unnecessarily long. For example Line 45 on Page 2 "Following their work ...... to address dynamic objects". Should be broken into two sentences. Similarly sentences on Line 88 and Line 96 need to be rewritten. Due to extending sentences too long, the grammar of the sentences become incorrect.  

4. The English language writing need to be reviewed overall by a native expert.  

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

1. Why in abstract, its mentioned as "depth estimation network" , why network here? what it signifies?

2. use first letter in capital for " generative adversarial network (GAN)"

3. auto hyphenation can be disabled in word

4. In abstract, its specified as "Our method", "our model". Use a name for model, instead of using "our". Also use "proposed" like "proposed LUMDE model". 

5. Why PoseNet and DepthNet - both needed?

6. Why Depth Estimation is necessary?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The concerns are adequately addressed. The paper is acceptable in the current form. 

Back to TopTop