GewelsJI / SINet-V2

Concealed Object Detection (SINet-V2, IEEE TPAMI 2022). Code is implemented by PyTorch/Jittor frameworks.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

What is the difference between RF and TEM?

lunaWU628 opened this issue · comments

It's a great job!

Question1: What is the difference between RF and TEM? In my opinion, RF and TEM are the same。

SINetv1
image
SINetv2
image

Question2: I want to use the TEM module, which part of the network structure works best? (Is it OK to put the TEM module behind the backbone?) )

This may be a simple question, but it is important to me, hope to get a response from you, thank you!

Hi, @lunaWU628

Thank you for raising such a good question here.

A1: As described in original paper, we just use RF-like module to extract texture cues. I refer you to the first part of section 4.2: "This motivates us to use the TEM [63] to incorporate more discriminative feature representations during the searching stage (usually in a small/local space)."

A2: I really suggest you use RF-like module to extract multi-scale information from backbone's features. It usually improves the model performance.

I hope these answers help you a lot.

Best,
Ge-Peng.

Thank you very much for your quick answer, this is very helpful to me.

Best wishes for you!