Spatial–Temporal Analysis-Based Video Quality Assessment: A Two-Stream Convolutional Network Approach

Round 1

Reviewer 1 Report

The submitted manuscript discussed about a two -stream convolutional neural network for video quality assessment (VQA) method. Extracting spatial and temporal features, and constructing a joint fusion system are the main contributions of the manuscript. However, I have following comments: 

Specific Comments: 

1) It is better to demonstrate the distribution of features in extraction and fusion stage, respectively.

2) What are the experimental results of quality scores. How they impact to the PLCC and SROCC. Do they have any useful connection. It is better to describe in the manuscript. 

Author Response

Thanks for your comments. We have carefully revised the manuscript. Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

At the end of subparagraph 2 authors included a proposed diagram, which they did not refer to in the text. In subparagraph 3authors will describe in more detail the scheme, which remains in the second subparagraph. I strongly recommend adding a logical conclusion to subparagraph 2 and referring to the drawing in it, or moving the drawing to the third subparagraph.

Having thoroughly studied the article, I would like to recommend authors add a Discussion section, or add one paragraph of text to the Conclusion. In this paragraph, the authors would pay attention to the areas of application of the proposed model, i.e. where and how it can be used. The general contribution to the mechanism for assessing video quality is clear, but it is important to explain to the average reader in which specific processes the proposed model can be applied in practice.

Author Response

Thanks for your comments. We hav carefully revised the manuscript. Please see the attachment.

Author Response File: Author Response.docx