Comments (13)
We used pose and shape parameters generated by the MoSh method. However, due to license implications, this data is not public. You could try to fit SMPL to the 3D keypoints of the dataset. This would probably be less accurate, but it should give you reasonable results.
from spin.
No MoSh results are available for MPI-INF-3DHP. In that case, we fit SMPL to keypoints. Since ground truth is not very accurate, we prefer to use OpenPose predictions and fit SMPL to them for all viewpoints simultaneously.
from spin.
@geopavlakos @BICHENG @liwenssss @vigorbird @GuiZhaoyang
where can i get the json file in the code
json_file = os.path.join(openpose_path, 'coco', imgname.replace('.jpg', '_keypoints.json'))
from spin.
This file includes the data for Human3.6M and is similar to the other .npz files we released. Besides image names, scale/center and 2D/3D/openpose joints, it also includes pose and shape parameters for the Human3.6M dataset. Due to the dataset license, we cannot release it, but if you have the data, it should be straightforward to produce it, if you also take a look to the other dataset preprocessing scripts we provide.
from spin.
Could you please add this h36m script to this repository?
from spin.
The script is already in SPIN/datasets/preprocess/h36m_train.py. I updated it to include the reading of the openpose predictions, but you will still need to read and include the SMPL pose and shape parameters in the .npz file (in case you have access to them).
from spin.
if I have h36m dataset , how can I generate pose and betas paramers?
from spin.
Do you generate pose and shape parameters of MPI-INF-3DHP in the same way?
from spin.
from spin.
The code for the MoSh approach is not publicly available. You could refer to the description of the paper for more details.
from spin.
@geopavlakos hi,Could I know how to fit SMPL to the 3D keypoints of the h36m dataset?Because I have not the Mocap annotation but I have the 3D keypoints annotation.May I know the code to to fit SMPL to the 3D keypoints?
from spin.
@GuiZhaoyang 兄弟留个邮箱,有偿求human3.6m mosh数据
from spin.
No MoSh results are available for MPI-INF-3DHP. In that case, we fit SMPL to keypoints. Since ground truth is not very accurate, we prefer to use OpenPose predictions and fit SMPL to them for all viewpoints simultaneously.
Hi @geopavlakos et al., Could you kindly release openpose detection of these data please? Thx & best,
from spin.
Related Issues (20)
- final fits npz file link not working HOT 5
- Differnece between unaligned and aligned H3.6m moshed data?
- Project
- How can I generate openpose json for MPII?
- Questions about data augmentation
- RuntimeError: size of dimension does not match previous size, operand 1, dim 2
- Some error I got HOT 2
- pck
- AssertionError: Path ./body_models/smpl/SMPL_NEUTRAL.pkl does not exist! HOT 2
- Some problems about body shape
- The formula for camera translation
- About reimplement HOT 1
- About performance HOT 2
- About the parameters of SMPL poses and camera pose
- Error Omesa
- How to use demo.py to get more result? Like Mesh result ...
- About traing results
- About the evaluation of H36M
- demo.py freezes on model prediction
- `FileNotFoundError` in colab notebook HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from spin.