Next Article in Journal
Molecular Hydrogen: The Postharvest Use in Fruits, Vegetables and the Floriculture Industry
Next Article in Special Issue
Estimates of Internal Forces in Torsionally Braced Steel I-Girder Bridges Using Deep Neural Networks
Previous Article in Journal
Ultrasound-Assisted Extraction of an Extract with Dermatological and Nosocomial Activity from Agave nuusaviorum, a Mexican Endemic Plant
Previous Article in Special Issue
Data-Driven Fault Detection and Diagnosis: Challenges and Opportunities in Real-World Scenarios
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 12
Issue 20
10.3390/app122010445
Review Report
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

Binary Dense SIFT Flow Based Position-Information Added Two-Stream CNN for Pedestrian Action Recognition

Appl. Sci. 2022, 12(20), 10445; https://doi.org/10.3390/app122010445
by Sang Kyoo Park 1, Jun Ho Chung 1, Dong Sung Pae 2,* and Myo Taeg Lim 1,*
Reviewer 1:
Nina Danilina
Reviewer 2:
Lyuchao Liao
Reviewer 3: Anonymous
Appl. Sci. 2022, 12(20), 10445; https://doi.org/10.3390/app122010445
Submission received: 26 September 2022 / Revised: 10 October 2022 / Accepted: 13 October 2022 / Published: 17 October 2022
(This article belongs to the Special Issue Artificial Intelligence and Complex Systems Analysis in Transportation and Maintenance)

Round 1

Reviewer 1 Report

The article is written on a very progressive topic of the use of artificial intelligence to solve the problems of a safe environment for pedestrians. The application of Binary Dense SIFT Flow based Position-Information Added Two-Stream CNN for Pedestrian Action Recognition is a unique technique that can predict pedestrian behavior in the driving environment. The authors described in detail and clearly the entire procedure for obtaining information and provided general recommendations for their use. In my opinion, this article lacks only the Discussion section, in which it is necessary to highlight the issues of applying the presented methodology to solve specific problems, as well as to highlight the issue of possible changes in pedestrian behavior depending on external factors. Basic questions: 1. How to understand that the obtained results of the pedestrian behavior forecast can be used in the medium term period and this behavior won`t be changed? 2. Whether the results will reflect a specific situation at the intersection or crossing. 3. Under what conditions the results of applying the method can be distributed to the  urban planning policy or street design.  As a whole, my assessment of the article is high.

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

This work aims to a interesting and challenging topic; however, there are still some respects that could be improved. For example, the contributions need to be refined, and more details about the model need to be  provied.  Additionally, more latest relevant references should be cited.

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper presents a CNN supported system for pedestrians behaviour on the road. The system attemps to predict the pedestrians next moves basing on the current observations.

The study is very important and innovative. In particular, the results can be applied to the further design of autonomous vehicles.

The research is very well designed, well documented and presented very clearly.

In my opinion, the paper can be published in the Journal as is

 

Author Response

Thank you for your valuable comments. 

Back to TopTop