Code for our CVPR 2020 (ORAL) paper - TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style.
This repository contains training and inference code for the following paper:
TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style
Chaitanya Patel*, Zhouyingcheng Liao*, Gerard Pons-Moll
CVPR 2020 (ORAL)
[ArXiv] [Project Website] [Dataset Repo] [Oral Presentation] [Results Video]
Cite us if you use our model, code or data:
@inproceedings{patel20tailornet,
title = {TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style},
author = {Patel, Chaitanya and Liao, Zhouyingcheng and Pons-Moll, Gerard},
booktitle = {{IEEE} Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {jun},
organization = {{IEEE}},
year = {2020},
}
python3
pytorch
scipy
chumpy
psbody.mesh
Code works with psbody.mesh v0.4 , pytorch >= v1.0 , chumpy v0.7 and scipy v1.3 .
global_var.py
file accordingly.global_var.py
.
run_tailornet.py
and run it to predict garments on some random inputs. You can play with
different inputs. You can also run inference on motion sequence data.python run_tailornet.py render
. (Blender 2.79 needs to be installed.)... evaluated using evaluate
function in utils/eval.py
.
garment_class | gender | TailorNet Baseline | TailorNet Mixture Model |
---|---|---|---|
old-t-shirt | female | 11.1 | 10.7 |
t-shirt | female | 12.6 | 12.3 |
t-shirt | male | 11.4 | 11.2 |
shirt | female | 14.2 | 14.1 |
shirt | male | 12.7 | 12.5 |
pant | female | 4.7 | 4.8 |
pant | male | 8.1 | 8.1 |
short-pant | female | 6.8 | 6.6 |
short-pant | male | 7.0 | 7.0 |
skirt | female | 7.7 | 7.8 |
global_var.py
, especially LOG_DIR where training logs will be stored.trainer/base_trainer.py
(or pass them via command line)
and run python trainer/base_trainer.py
to train TailorNet MLP baseline.python trainer/lf_trainer.py
to train low frequency predictor and trainer/ss2g_trainer.py
to
train shape-style-to-garment(in canonical pose) model.python trainer/hf_trainer.py --shape_style <shape1>_<style1> <shape2>_<style2> ...
to train pivot high
frequency predictors for pivots <shape1>_<style1>
, <shape2>_<style2>
, and so on. See
DATA_DIR/<garment_class>_<gender>/pivots.txt
to know available pivots.models.tailornet_model.TailorNetModel
with appropriate logdir arguments to do prediction.In the paper, we report inference time to be 1-2 ms per frame(depending upon garment) which is averaged inference time over the batch of 21 samples(20-40 ms per batch). Apologies for the ambiguity. Running each sample separately takes almost same time as batch - around 20 ms per frame for all garments. However, note that TailorNet has 21 independent MLPs, so we believe that faster inference time is possible if MLPs are configured to run in parallel on GPU cores.
smpl_lib
library taken from his MultiGarmentNet
repo's lib folder.For any doubt or concert about the code, raise an issue on this repository.