All Directional Search Motion Estimation Algorithm

Round 1

Reviewer 1 Report

In this paper, a search pattern is proposed to improve the performance of block matching estimation algorithms.

The proposed algorithm was verified by comparing it with existing algorithms in terms of ANOB.

Overall, I think that the paper is well-organized, and the results that can verify the performance are presented sufficiently.

Some comments are provided to improve the readability of the thesis.

When comparing ANOB performance for sample videos, it is desirable to display the same algorithm with the same color.

In Figure 9, ANOB of ADS generally shows the same trend as other algorithms, but (c) shows the performance of ADS with different tendencies. I think further explanation is needed.

Author Response

Dear Editor and Reviewers,

We would like to thank you for the careful and thorough reading of this manuscript mentioning thoughtful and supportive comments and constructive suggestions, which help us to improve this manuscript's quality. Please see below, in blue, our detailed response to comments. (Comment is in Red color)

 

Reviewer 1

In this paper, a search pattern is proposed to improve the performance of block matching estimation algorithms.

The proposed algorithm was verified by comparing it with existing algorithms in terms of ANOB.

Overall, I think that the paper is well-organized, and the results that can verify the performance are presented sufficiently.

Some comments are provided to improve the readability of the thesis.

Comment-1: When comparing ANOB performance for sample videos, it is desirable to display the same algorithm with the same color.

Note: The sequence of figure numbers 9 and 10 are corrected to Figure 8 and 9. We request the reviewers observe the modification.

Author’s Response-1: Modified as suggested. Now, the Figure displays the same algorithm (ANOB and others) with the same color

Comment-2: In Figure 9, ANOB of ADS generally shows the same trend as other algorithms, but (c) shows the performance of ADS with different tendencies. I think further explanation is needed.

Note: The sequence of figure numbers 9 and 10 are corrected to Figure 8 and 9. We request the reviewers observe the modification.

Author’s Response-2: The video sequence of Figure 9 (c) (Actually it is now Figure 8 (c) in the manuscript) has complex motions and particularly the proposed ADS demands less computation. The variation in ANOB of ADS is also small. So, tendency is not identifiable even it shows over a small range.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper proposes an all directional search (ADS) pattern that searches for the best block from all possible directions and shows that the proposed ADS algorithm is superior to other motion estimation algorithms. The proposed ADS algorithm shows excellent performance, and I believe the author's proposed algorithm is quite useful for motion estimation. We believe that this paper should be published, but there are a few corrections that need to be made: 

1. In Figures 9 and 10, the meaning of the "FRAME NUMBER" axis needs to be explained. Perhaps the meaning of this axis is not mentioned in the text. 

2. In the "Funding" section, the double quotation marks need to be removed. This is the same for the "Conflicts of Interest" section.

Author Response

Dear Editor and Reviewer,

We would like to thank you for the careful and thorough reading of this manuscript mentioning thoughtful and supportive comments and constructive suggestions, which help us to improve this manuscript's quality. Please see below, in blue, our detailed response to comments. (Comment is in Red color)

 Reviewer 2

This paper proposes an all directional search (ADS) pattern that searches for the best block from all possible directions and shows that the proposed ADS algorithm is superior to other motion estimation algorithms. The proposed ADS algorithm shows excellent performance, and I believe the author's proposed algorithm is quite useful for motion estimation. We believe that this paper should be published, but there are a few corrections that need to be made:

Comment-1: In Figures 9 and 10, the meaning of the "FRAME NUMBER" axis needs to be explained. Perhaps the meaning of this axis is not mentioned in the text.

The sequence of figure numbers 9 and 10 are corrected to Figure 8 and 9. We request the reviewers observe the modification.

Author’s Response-1: In these figures, the X-axis represents the sequence number of each frame in the video sequence. The following text below in bold has been added in the paper at line number 242.

 

In these figures, the X-axis represents the sequence number of each frame in the video sequence.

 

Comment-2. In the "Funding" section, the double quotation marks need to be removed. This is the same for the "Conflicts of Interest" section.

Author’s Response-2: Modified as suggested by the reviewer

 

Author Response File: Author Response.pdf