Comments (2)
The flickering could be down to poor interpolation and or a frame rate out of sync with the pose video. I think it would be best to lock the FPS to the pose video and then allow the user to choose the length in seconds, so long as it doesnt exceed the length of the pose video.
from moore-animateanyone.
Hi, @yihong1120 , thanks a lot for your valuable suggestions. During the reproduction process, we indeed encountered the issues you mentioned. Through filtering and cleaning of the training data, as well as employing some training tricks (we will soon release the training code), we observed a mitigation of such issues.
Certainly, we are currently exploring some of our own optimization measures to further alleviate these issues. We will also consider the valuable suggestions you provided, and you are welcome to experiment with improvements once we release the training code. We look forward to further collaboration and discussions in the future.
from moore-animateanyone.
Related Issues (20)
- stage2训练卡0显存泄露,爆显存
- can anyone explain the classifier-free guidance in referencenetattention? HOT 1
- how to realize pose2img ? can i modify pose2vid based on pipeline_pose2img.py???
- Set gradient_accumulate_steps > 1 will occur to accelerate.backward() twice error HOT 1
- additional dataset were added?
- Will you fix face quality later? HOT 2
- quantitative results
- inference for stage1 HOT 1
- Inconsistency of classifier-free guidance between training and testing. HOT 1
- Question about Finetue MotionModel
- how can i return just inference video HOT 2
- Why the height in train stage 1 and train stage 2 are different?
- Question about training memory usage HOT 2
- question about inference parameters
- Is there any difference between diffusers unet_2d_condition and yours unet_2d_condition?
- Misimplementation of Spatial-attention? HOT 2
- Inference Too Buggy
- During both stage 1 and stage 2 training, the training stops unexpectedly for unknown reasons.
- How to reduce the size of the resulting video (while maintaining high quality) HOT 2
- Some top layer parames of reference_unet don't need grad, what is the reason? 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 moore-animateanyone.