Skip to content

TalkUHulk/pixel2style2pixel-mobilenetv3

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

pSp with MobileNetV3


The demo of different styles images of pSp-mbv3 framework.

Description

Re-implementation of pSp that use mobilenet-v3 and stylegan2-256p. Then mix different styles by stylegan2 and convert to openvino.

Citation

This code is heavily based on pixel2style2pixel and stylegan2-pytorch. Thanks eladrich and rosinality so much to make his work available 🙏🙏🙏

Pretrained Models

链接: https://pan.baidu.com/s/1CjJIvhUXk0f9ONPANK-fSg 密码: pjl8

Training Log

Tensorboard


The loss of training.

inversions


iteration 0 and iteration 100000.

Test

python scripts/inference.py \
--exp_dir=./sample \
--checkpoint_path=./ckpt/best_model.pt \
--data_path=./test_images \
--test_batch_size=4 \
--test_workers=2 \
--couple_outputs

More

Inference on openvino and more demo

About

Re-implementation of pSp that use mobilenet-v3 and stylegan2-256p

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published