Detect Orientation of Symmetric Objects from Monocular Camera to Enhance Landmark Estimations in Object SLAM

Round 1

Reviewer 1 Report

Manuscript ID: applsci-2185082

Title: The Enhancement of Landmark Estimation in Object SLAM by Orientation Detect of Symmetric Objects from Monocular Camera

Authors: Fang et al.

In this manuscript, authors proposed a strategy that first extracts the orientations of those symmetric human-made objects in a single frame, and then implements the orientations results as back-end constraint factor of the ellipsoid landmarks to improve the performance of Object SLAM system with ellipsoid landmarks. Authors claimed that the proposed orientation detection method  achieves lower orientation error in most tested datasets compared with the baselines, and meanwhile improves the accuracy of mapping.

In my opinion, the novelty of the presented work is good. And the manuscript is well presented and organized. In addition, the topic fits the scope and aims of the Applied Sciences journal. The paper can be accepted after minor revision as follow:

1-    The title includes abbreviation. Please avoid the abbreviation in the title

2-    The size of all figures in the manuscript should be enlarged. So details inside figures cab be easily readable.

3-    All equations used in this work should have references.

4-    Results of this work should be compared with other reported results. For example, the calculated  Translation, Rotation and Shape Error.

5-    Kindly check the English through whole manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors present enhancement of Landmark Estimation in Object SLAM by Orientation Detect of Symmetric Objects from Monocular Camera . The study is interesting.  In general, the main conclusions presented in the paper are supported by the figures and supporting text. However, to meet the journal quality standards, the following comments need to be addressed.

•           Abstract: Should be improved and extended. The authors talk lot about the problem formulation, but novelty of the proposed model is missing. Also provided the general applicability of their model. Please be specific what are the main quantitative results to attract general audiences.

•           The introduction can be improved. The authors should focus on extending the novelty of the current study. Emphasize should be given in improvement of the  model (in quantitative  sense)  compared to   existing  state-of-the art models.

•           More details about network architecture and complexity of the model should be provided.

•           what about comparison of the result with current state-of-the art models?  Did authors perform ablation study to compare with different models?

•           What are the baseline models and benchmark results? The authors may compared the result with existing models evaluated with datasets

•           Conclusion parts needs to be strengthened.

•           Please provide a fair weakness and limitation of the model, and how it can be improved.

•           Typographical errors: There are several minor grammatical errors and incorrect sentence structures. Please run this through a spell checker.

·       Discussions of relevant literature could be further enhanced, which can help better motivate the current study and link to the existing work, for example,  relevant deep learning references on object detections   ( see :

-Neural Networks 2022 https://doi.org/10.1016/j.neunet.2022.05.024

-Ecol. Informatics 2022 https://doi.org/10.1016/j.ecoinf.2022.101919

-arXiv (2022)  https://doi.org/10.48550/arXiv.2210.04252

Hence they should be briefly discussed in the related work section.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

In this paper, a method to utilize symmetry of human-made objects to enhance landmark estimation for monocular Object SLAM system is proposed. The mapping rotation angle required by the camera coordinate system is found by constructing a cost function to determine the symmetry axis direction of the object. This paper has certain application value. However, there are several problems questions to be answered in the following aspects:

1. How does the weight been set in equation (2), and what is the effect of changing the weight value?

2. Please pay attention to the complete annotation of abbreviations. For example, the interpretation of SLAM in the abstract needs to be refined.

3. It is recommended to add a schematic diagram of the directional relationship to be labeled in section 3.2.

4. What does the variable K in equation (4) represent? Please further explain it.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The revised version is now suitable for publication.