search

JaehaKim97 / SR4IR

[CVPR 2024] Official implementation of "Beyond Image Super-Resolution for Image Recognition with Task-Driven Perceptual Loss"
45 stars 2 forks source link

咨询 #4

Open MaoFuyou opened 1 month ago

MaoFuyou commented 1 month ago

想问一下，如果我做的是三维图像的模态转换，那么VGG模型应该怎么加载预训练权重呢？因为是不是没有VGG的三维权重呢？

JaehaKim97 commented 1 month ago

Hi @MaoFuyou,

I'm sorry, but I can not understand Chinese. :cry: Could you repeat the question in English?

MaoFuyou commented 1 month ago

Thank you .In the context of three-dimensional image modality transformation, how should one load pre-trained weights for the VGG model? Is it not the case that there are no pre-trained weights available for three-dimensional VGG models?

JaehaKim97 commented 1 month ago

As far as I know, the VGG model is not trained for 3D images. So, pre-trained weights for the three-dimensional VGG model do not exist. Maybe you can use some other pre-trained 3D-image classification networks as an alternative.