Code repositoty for the paper:
Out-of-Domain Human Mesh Reconstruction via Dynamic Bilevel Online Adaptation
Shanyan Guan, Jingwei Xu, Michelle Z. He, Yunbo Wang, Bingbing Ni, Xiaokang Yang
We focus on reconstructing human mesh from out-of-domain videos. In our experiments, we train a source model (termed as BaseModel) on Human 3.6M. To produce accurate human mesh on out-of-domain images, we optimize the BaseModel on target images via DynaBOA at test time. Below are the comparison results between BaseModel and the adapted model on the Internet videos with various camera parameters, motion, etc.
DynaBOA has been implemented and tested on Ubuntu 18.04 with python = 3.6.
Clone this repo:
git clone https://github.com/syguan96/DynaBOA.git
Install the requirements using miniconda
:
conda env create -f dynaboa-env.yaml
Download required file from this link. Then unzip the file and rename it to data
folder. Additionally, download sampled human 3.6M data from this link and unzip it to data/retrieval_res
Download Human 3.6M using this tool, and then change the corresponding path at Line 700 in dynaboa.py
.
bash run_on_3dpw.sh
Method | Protocol | PA-MPJPE | MPJPE | PVE |
---|---|---|---|---|
SPIN | #PS | 59.2 | 96.9 | 135.1 |
PARE | #PS | 46.4 | 79.1 | 94.2 |
Mesh Graphormer | #PS | 45.6 | 74.7 | 87.7 |
DynaBOA (Ours) | #PS | 40.4 | 65.5 | 82.0 |
- DynaBOA for MPI-INF-3DHP and SURREAL
- DynaBOA for the internet data.