Aircraft Carrier Pose Tracking Based on Adaptive Region in Visual Landing

Round 1

Reviewer 1 Report

In this research article, the authors proposed a scale-adaptive method for aircraft carrier pose tracking. The proposed method is extensively discussed in great detail. The authors conducted comparative experiments with method from competiters on synthetic and real image sequence respectively to verify the efficiency and accuracy of the proposed pose tracking method. However, the authors need to address the following queries :

1.     The authors mentioned that the proposed method requires an initial pose and the intial pose is obtained by PnP in the experiment on real image sequence. But the intial pose cannot always be obtained with such precision, and the authors should provide the sensitivity analysis of the initial pose with extra experiments.

2.     The background of the real image sequence is too complex to simulate the real landing operation while the aircraft carrier on the sea. Is this the reason why other methods perform poorly? The authors should add an extra experiment based on a real image sequence with a clean background to explain this. And the adaptability of the proposed method to complex backgrounds should be considered as an additional advantage.

3.     The authors conducted comparative experiments with other similar methods and presented the results in figures. The authors should align the color of the results corresponding to each method in the different figures (i.e. one color for one method, including reprojected results and pose curves), to enhance the intuitiveness of the results.

4.     There are several typographical errors (e.g. line 219). The authors should proofread the entire article.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Editing notes:

line 392 Move the caption to the previous page

line 219  . -> forward to the line above

Detailed comments:

The paper is very interesting and presents the problem in a comprehensive way. The authors present the proposed algorithm, compare it with two methods. The results obtained clearly indicate the great potential of the proposed algorithm. Simulation studies on a scale model confirm the advantages of the proposed algorithm.

The paper is carefully edited, the chapters are arranged logically and consistently. The conclusions are supported by the presented research results. Experiments conducted on synthetic and real image sequences confirm that the proposed method can accurately and reliably track the position of an object.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

 

This paper considers a new robust aircraft carrier pose tracking algorithm based on scale adaptive local region, which achieves better performance than existing algorithms. Also, a new updating mechanism of local region model parameters is proposed, considering the target scale change in the image.

The paper is well structured and the purpose of the paper is well explained. The experimental results demonstrate the better performances of the proposed algorithm to the existing approaches. The citations are adequate.

The following remarks could be made.

The theoretical results are not presented with sufficient details. The Subsection 2.2 is not clearly written. The lot of equations are given without explanation or citations, especially equations (7)-(17). Authors should provide better explanation or additional citations, to improve readability of this Subsection.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf