An Object Detection Method Based on Feature Uncertainty Domain Adaptation for Autonomous Driving

Round 1

Reviewer 1 Report

Considering automatic driving vehicles as reality of the near future, detection of surrounding objects must be treated as very important, especially from safety point of view. Studies on this subject are important for predictions of important tasks to be covered by the autonomous driving vehicles.

Related works is well represented. Research methods are good explained using both mathematical modelling and experimental analysis.

Paper conclusions can be improved to reflect all the results obtained through the study and to increase the paper value.

Some figures are recommended to be improved in quality, like Figure 1, Figure 5 and Figure 9.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The subject treated by the authors is a topical one. Researchers and companies are continuously adapting the systems on the vehicles and are concerned with improving the perception time of objects in autonomous driving.

Even if the image processing and analysis method is widely used, in this paper, the authors propose an adaptive object detection algorithm based on the uncertainty of the characteristics depending on the environmental changes.

The documentation is comprehensive and shows a thorough knowledge of the methods of detection and analysis of objects, many works being reviewed.

Remarkable all the analysis made by the authors in different weather conditions and urban landscapes. The results obtained are remarkable.

 In conclusion the paper is  good in terms of scientific contribution.

Author Response

We sincerely appreciate your interest in our work and extend our gratitude for dedicating your time to review our manuscript.