search

zwx8981 / UNIQUE

The repository for 'Uncertainty-aware blind image quality assessment in the laboratory and wild' and 'Learning to blindly assess image quality in the laboratory and wild'
Apache License 2.0
127 stars 10 forks source link

bilinear pooling layer #6

Closed pencilzhang closed 3 years ago

pencilzhang commented 3 years ago

Hi, you proposed to use bilinear pooling instead of first-order average pooling in the original resnet, as mentioned in the paper. However, I found the average pooling is kept in the implementation of resnet (BaseCNN.py). A bilinear pooling layer is added on top of that, no?

zwx8981 commented 3 years ago

Yes, I didn't remove the original 1000-dim fc layer. But it will not be updated during the training and has no effect on the testing.