Comments (5)
hi Yuanzhi, thanks for your interest in the work.
The benefits of writing the score as
- You can visualize
$D$ .$D$ is a (potentially blurry) image. This is helpful for debugging and understanding. - It is clear that
$D - x$ is mean-shift. It is more intuitive. -
$D - x$ is also the gradient of a l2 reconstruction loss. This is helpful for understanding what's happening. When doing 3D optimization, 2D diffusion guidance is providing 1-step reconstruction targets.
And sry about the earlier deleted response. Since you are referring to score at time
from sjc.
Also
The consensus on denoiser parametrization now leans towards
Since in the end it's all denoising, it seems easier to treat them all as
from sjc.
Hi Haochen, thank you so much for elaborating on the advantages of writing the score as
Here I have a follow-up question regarding the difference between
from sjc.
Scaling down increases the density. The denominator is
If there is no trajectory scaling then
DDPM uses scaling to cap the variance (which, only in hindsight, appears unnecessary). VESDEs are easier to solve and the formula tends to be simpler.
from sjc.
Thank you so much for your clarification.
from sjc.
Related Issues (20)
- Different angles in SD HOT 1
- The next SD version could speed up the generation of 3d models ? "Stable Diffusion will generate 30 images per second instead of one image in 5.6 seconds" HOT 1
- Drop the duplicate running process in the queue on Hugging Face
- Details about the eq19 HOT 2
- About bs=1 and the use of the Monte-Carlo estimate HOT 4
- question of the provement of the formula17 HOT 3
- Question about train_eye_with_prompts HOT 3
- scripts for figure 4 HOT 2
- Can you provide the rendered videos for better comparison?
- About OOD issues. HOT 2
- Questions about center depth loss in the paper HOT 5
- Where is the jacobian implementation for the backpropogation?
- Bibtex not up to date HOT 1
- noisy_x scaling before diffusion models HOT 1
- Query : 4 channels in nerf output HOT 5
- Question about paper
- How to train the model with a new dataset? HOT 1
- Multi-face Janus issue HOT 7
- Colab ModuleNotFoundError: No module named 'torchtext.legacy' HOT 2
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 sjc.