DFDNet

DFDNet copied to clipboard

I understand that I'm supposed to edit gpu_ids to -1 but nobody says in which file. i should note that i am using dfdnet as a mve plugin as i...

Pluviophille

I followed the video, the program runs, but I got all the pictures with a covered face, and the face is something like a heatmap, can't understand.

cliobeach

Can I skip step 1 (crop and align faces)?

2

Hello, So, I have like 82700 pictures of cropped and aligned faces. Wait for step 1 to be done is a bit long so I am wondering if step 1...

xamxixixo

Way to Use GPU for Steps 1-2 and CPU for 3-4 (Link to Google Drive of .pys is included below)

10

First, I love this program. It is excellent at what it does and works better than any other I have seen. Due to the issues with running DFDnet on a...

zarc70

DFT bloack seems only can training on batchsize=1 ?

1

such as : get the part location of the vggFeat. F.inteplate, adaptivate_instance_norm, the feature matches, these process only can run on batch size=1 ?

miaoYuanyuan

About update face dictionary

6

Dear xiaomingLi, You have done a great job! Thanks for sharing! I am trying to update the face dictionary by following your code and the instructions on your paper. However,...

Nise-2-meet-U

I am testing the restoration results based on pretrained DFDNet and get a weird results on side face as the network's input. The following is the input and output: We...

ZhangYuef

Which discriminator architecture did you use to train the network?

funnylittleman

For perceptual loss do you use features obtained by VGGFeat class(same features as in main model) or some other model? In paper it is written that you use first 4...

funnylittleman

I get this error when trying --gpu_ids 0,1 "There is an imbalance between your gpus. You may want to exclude gpu 1 which has less than 75% of the memory...

pointave