An Underwater Human–Robot Interaction Using a Visual–Textual Model for Autonomous Underwater Vehicles
, Yu Jiang 1,2,*, Hong Qi 1,2, Minghao Zhao 1, Yuehang Wang 1, Kai Wang 1 and Fenglin Wei 1
Round 1
Reviewer 1 Report
I suggest you title the last part of the article as "summary"/"conclusions"
"Author Contributions" should be a separate section/chapter
The authors focused on communication in one direction: from the diver to the robot. I would expect at least one paragraph of commentary, e.g. in a summary of how the proposed approach can be used in robot-to-diver communication.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper proposes an visual-textual model for underwater hand gesture recognition (VT-UHGR). The VT-UHGR model encodes the underwater diver’s image as visual features, the category text as textual features, and generates visual-textual features through multimodal interactions. The following points need to be carefully considered during the revision.
-The datasets taken for the study are very minimum. The authors are advised to considered all the possible combinations of hand gesture.
- The results discussion requires all other parameters also not only the accuracy. The loss function results examination need to be presented.
- The validation of the proposed method requires much more analysis with different data sets. Moreover the comparative study is weak.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors addressed my concerns. Paper can be accepted for publications
