Awesome

PINet_PG

Code for our PG paper Human Pose Transfer by Adaptive Hierarchical Deformation

This is Pytorch implementation for pose transfer on DeepFashion dataset. The code is extremely borrowed from Pose Transfer. Thanks for their work!

Requirement

conda create -n tip python=3.6
conda install pytorch=1.2 cudatoolkit=10.0 torchvision
pip install scikit-image pillow pandas tqdm dominate 

Data

Data preparation for images and keypoints can follow Pose Transfer Parsing data can be found from baidu (fetch code:abcd) or Google drive

Test

You can directly download our test results from baidu (fetch code: abcd) or Google drive.<br> Pre-trained checkpoint can be found from baidu (fetch code: abcd) or Google drive and put it in the folder (-->checkpoints-->fashion_PInet_PG).

Test by yourself <br>

python test.py --dataroot ./fashion_data/ --name fashion_PInet_PG --model PInet --phase test --dataset_mode keypoint --norm instance --batchSize 1 --resize_or_crop no --gpu_ids 0 --BP_input_nc 18 --no_flip --which_model_netG PInet --checkpoints_dir ./checkpoints --pairLst ./fashion_data/fasion-resize-pairs-test.csv --which_epoch latest --results_dir ./results

Citation

If you use this code, please cite our paper.

@article{pinet,
	author = {Zhang, Jinsong and Liu, Xingzi and Li, Kun},
	title = {Human Pose Transfer by Adaptive Hierarchical Deformation},
	journal = {Computer Graphics Forum},
	volume = {39},
	number = {7},
	pages = {325-337},
	year = {2020}
}

Acknowledgments

Our code is based on Pose Transfer.