Infrared and Visible Image Homography Estimation Based on Feature Correlation Transformers for Enhanced 6G Space–Air–Ground Integrated Network Perception

Round 1

Reviewer 1 Report

The reviewer would like to thank the authors for this thoughtful manuscript. This work has good potential. The authors are requested to put in some additional efforts to improve the quality of this manuscript. 

Related Work

This section has to be merged with the Introduction as a part of the literature. Generally such a section is not present.

Co-Registration in Interdisciplinary Image Processing Applications

The authors are requested to discuss the cross domain application of the proposed technique for coregistration of images and cite the following interdisciplinary articles by highlighting the importance of the proposed technique for the applications addressed in the articles.

-Shugar et al, A massive rock and ice avalanche caused the 2021 disaster at Chamoli, Indian Himalaya, Science, 2021.

-Muhuri et al., Glacier surface velocity estimation using Stokes vector correlation. IEEE 5th Asia-Pacific Conference on Synthetic Aperture Radar (APSAR), 2015.

-Schmah et al., 2010. Comparing computational methods for longitudinal fMRI studies. Neural Computation.

Fig. 1

This figure is poorly presented. The authors are requested to enhance the visualization of this figure. The other figures have better clarity.

Fig. 6

This figure is confusing. There are slots that are missing and the reader has to find the reason in the text. Is it possible to present it in a better manner?

Discussion

This section is missing in the paper. Please include this section before the Conclusion.

Conclusion 

The authors are requested to list the key contributions in this section. At the moment the section is not detailed enough.

A grammatical check is recommended.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This study proposed a feature correlation transformer method for infrared and visible image homography estimation.

Comments/suggestions

·         Line number 414, do not start the sentence with “also”.

·         It is suggested to add a table of the experimental environment instead of text.

·         In this study synthetic dataset was considered for the experiment. It is suggested to test the proposed method on the real-world dataset.

·         Some of the recent relevant feature-based studies were not considered in the literature of the manuscript. A few of them are:

o   Forero, M. G., Mambuscay, C. L., Monroy, M. F., Miranda, S. L., Méndez, D., Valencia, M. O., & Gomez Selvaraj, M. (2021). Comparative Analysis of Detectors and Feature Descriptors for Multispectral Image Matching in Rice Crops. Plants, 10(9), 1791. https://doi.org/10.3390/plants10091791

o   Sharma, S. K., Jain, K., & Shukla, A. K. (2023). A Comparative Analysis of Feature Detectors and Descriptors for Image Stitching. Applied Sciences, 13(10), 6015. https://doi.org/10.3390/app13106015

o   Mukherjee, D.; Jonathan Wu, Q.M.; Wang, G. A comparative experimental study of image feature detectors and descriptors. Mach. Vis. Appl. 2015, 26, 443–466.

·         How the quantitative evaluation was done, is not clear.

·         More discussion can be included.

Grammatical mistakes were detected in some places.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

In this study, the authors propose a feature correlation transformer method, devised to offer explicit guidance for feature matching for the task of homography estimation between infrared and visible images. To validate the effectiveness of the newly proposed components, extensive experiments are conducted to demonstrate the superiority of the proposed method compared to existing methods in both quantitative and qualitative aspects. However, the following problem need to be improved.

(1) The English of paper is poor, please polish it again.

(2) Each variable in the equation should be given the meaning.

(3) Please give all parameters of the proposed method.

(4) How to train the model?

(5) More experiments should be given to show the advantage of the proposed method.

see the report

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

In this paper, the authors nicely describe the registration of two types of images: homography estimation of infrared and visible images.
They propose a feature correlation transformer method: they propose a patch feature which is used as the basic unit to calculate the correlation, a new inter-image attention mechanism to identify correlations between different modal images, they propose a feature correlation loss.
This will improve the image mapping between the sources and the target.
The theory is demonstrated by extensive experiments.
This document has a scientific and original character, therefore I recommend the publication for publication.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

All the comments/suggestions have been addressed by the authors.