Comments (8)
OK. I will try that. I tried static training before and everything looks good except that loss_regr_betas
has some extreme values once a while. I am curious about your training details. Did you train without run_smplify
at the beginning? Also, did you use the default loss weight here? It seems that shape_loss_weight
is 0? Also about openpose_train_weight
and gt_train_weight
, any suggest on this? Do you also suggest openpose_train_weight
as in the fitting stage?
from spin.
So just to be clear, the previous version of the code could still be using the extreme betas from the static fits. What I do is I check which fit has the lowest reprojection error and in the beginning of training it is almost certain that the static fits have lower error because the network has not started producing accurate shapes, i.e. good initializations for the optimization. This is one reason we enable the in-the-loop training later on.
from spin.
Hmm it looks like a learning rate issue. Can you try and train with a 10 times lower learning rate? Unfortunately Adam does not behave like SGD, so scaling the learning rate when increasing the batch size can have unexpected behavior.
Also try not to use --run_smplify in the beginning.
from spin.
The parameters that you mentioned are the ones used in the paper. I also inspected the code and I noticed that there is a small bug in the released version. Essentially I am not filtering out extreme betas in the static fits, only for the in the loop fits. I will do the modification now and I will push the changes.
from spin.
I think this should be the reason for the bumps in . I recalled in the paper you mentioned about the per vertices loss, i.e., shape loss in the code. Did you ever run the experiment on the shape loss? If so, did you disable it because of inferior performance?loss_regr_betas
from spin.
We mention in the supplementary material that the weight we chose for this loss is 0. The reason that we ended up not using this loss is that it is highly correlated with the other losses.
from spin.
Well, I just take a look at your commit b4c70d5. The above figures are run under run_smplify
. So I feel there may be other reasons for the bumps?
from spin.
@iammarvelous were you able to avoid this problem by reducing the learning_rate or enlarging the batch_size?
from spin.
Related Issues (20)
- 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
- Replace smpl model with smpl-x HOT 2
- demo.py blank result
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.