Frequency-Based Adaptive Interpolation Filter in Intra Prediction

Round 1

Reviewer 1 Report

the author has presented a VVC interpolation method that could obtain better performance than another method in the literature. Even though the proposed method can improve the average performance. There is one concern that there are cases when the proposed method can not produce good results. The author should discuss those cases, find out the reason and explain why the proposed method does not work as expected. 

Author Response

Attached please find coverletter-r1.docx.

Author Response File: Author Response.docx

Reviewer 2 Report

This paper proposes a fractional intra prediction method that adaptively conducts interpolation according to a frequency state of reference samples without any bit overhead in VVC. This manuscript would be helpful in the area of video coding technology as a way of possible improvement of the standard codec. The following comments need to be considered.

  1. It is required to modify equations and related explanations.
       a. Equations (1, 2) and (7) are redundant, and Equation (14) seems unnecessary to understand the proposed method.
       b. It is unclear whether the explanation for Equation (11) is to derive filter coefficients or to perform filtering.

  2. Minor corrections are needed in tables and figures.
    a. Types of methods in Table 10.
    b. Clarification on the x-axis in Figure 6.

  3. In Section 4, the proposed adaptive filter selection gives a meaningful coding gain. In addition, additional explanations of the advantages in other aspects could be added. For example, an analysis in terms of subjective quality could be added.

 

 

Author Response

attached please find coverletter-r1.docx

Author Response File: Author Response.docx