neuralchen / SimSwap

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

Pretrained network G has fewer layers; The following are not initialized: ['down0', 'first_layer', 'last_layer', 'up0'] #432

Open taaha opened 1 year ago

taaha commented 1 year ago

image

getting image like above when running !python test_one_image.py --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/

no error but this msg image

GoodFanxinran commented 1 year ago

the same, do you solve it?

KleinXin commented 1 year ago

me too

Xiujian-LIANG commented 1 year ago

+1

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/ @KleinXin @taaha

shidalao 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/ @KleinXin @taaha

it works, thanks

nghiap commented 1 year ago

Has anyone done a training set for Asian faces? if so may I please have a copy? My laptop is too old to do any training :( Please share, I am not doing any thing commercial... only mucking around with some photos for fun.