Real-Time Stereo Visual Odometry Based on an Improved KLT Method
Round 1
Reviewer 1 Report
I suggest the authors address the following comments:
1) Line 303, what do the authors mean by A prototype was written in MATLAB to prove the concept ? Please explain.
2) Line 276, how is the threshold angle of 0.00 rad defined?
3) Line 295, the sentence begins "In the following experiments ......", the authors explain if there were previous experiments to those described after this sentence.
4) Lines 310-322, the data mentioned in this paragraph do not coincide with those in Table 1, this is due to rounding error, authors are suggested to use the same format in both parts of the manuscript.
5) Line 334, the first row of Table 1 should be headed SVO + JAF + AVM and AKT + Disparity.
6) Line 322, the term ATK should be AKT.
7) Table 1, I suggest including a non-parametric means test of the data shown in this Table 1, in this way the results would have greater validity.
8) I suggest the authors include a more detailed conclusion on the comparison of the proposed system with the KITTI odometry benchmark.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript proposed stereo visual odometry (SVO) for real-time localization based on an improved Kanade-Lucas-Tomasi (KLT) feature tracker method. Here is the comment for the proposed work:
"Please add at least three more state-of-the-art methods (above 2020) to show the robustness of the proposed method."
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
In this paper, the Authors used non-scientific language. For example, "we present a novel SVO algorithm" in line 13, "we perform several comparative" in line 20, "Our approach is most similar to the SVO" in line 73, "We analyze the relationship between the motion" in line 77, "We perform a systematic evaluation" in line 88 and so on.
Authors should rewrite the paper with Passive Voice sentences used in Scientific language. The manuscript should be evaluated after this major language editing.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Thank you very much for answering all of my comments. Moreover, it would be much better if the proposed model also used C programming language, then the comparison would be in the same distribution. Good luck.
Author Response
Thank for you your suggestion. We totally agree that the proposed method should be implemented in C/C++ and optimized for real-time performance. Several field tests using the C programming language are actually under investigation. However, due to time constraints, we will demonstrate them in our future work.
Reviewer 3 Report
In revised version of the study most of the major language style problems are corrected by the Authors. But still some non-scientific language usage are found. For example, "In our case" in line 148, "In our implementation, we determine that" in line 150, "we indicate that" in line 189, "Therefore, in our case, we take as the" in line 290, "we compare the proposed method" in line 336 etc.
Authors should rewrite the paper with Passive Voice sentences used in Scientific language. The manuscript should be accepted after this minor language editing. Otherwise, the reviewer will reject this paper because these scientific writing problems are still found in the revised version.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
