neuralchen / SimSwap

An arbitrary face-swapping framework on images and videos with one single trained model!
Other
4.41k stars 872 forks source link

Strange Output #439

Closed superbunny38 closed 1 year ago

superbunny38 commented 1 year ago

Hello, thank you for sharing your work :)

I was trying to inference with your model using your given images and video, and the converted video seems very strange. I followed every code in colab, but the result shows as follows:

image

Also, I doubt if this line of warning is a clue for my problem: Pretrained network G has fewer layers; The following are not initialized: ['down0', 'first_layer', 'last_layer', 'up0']

I'll wait for your reply, thank you! :)

Xiujian-LIANG commented 1 year ago

I solved that. The first test file lack a hyperparameters. Just add the --crop_size 224/512 after py command. Such as python test_one_image.py --crop_size 224 --name people --Arc_path arcface_model/arcface_checkpoint.tar --pic_a_path crop_224/6.jpg --pic_b_path crop_224/ds.jpg --output_path output/ @superbunny38

superbunny38 commented 1 year ago

Oh! thank you so much :D