multimodalart HF staff commited on
Commit
7b9597e
1 Parent(s): bf8eb88

Update index.html

Browse files
Files changed (1) hide show
  1. index.html +2 -2
index.html CHANGED
@@ -263,14 +263,14 @@
263
  <p>
264
  Utilizing T2I models for editing real images is usually done by inverting the sampling
265
  process to identify a noisy xT that will be denoised to the input image x0.
266
- We draw characteristics from edit friendly DDPM inversion [] and propose an efficient
267
  inversion method that greatly reduces the required number
268
  of steps while maintaining no reconstruction error.
269
  DDPM can be viewed as a first-order
270
  SDE solver when formulating the reverse diffusion process as an SDE. This
271
  SDE can be solved more efficiently—in fewer steps—
272
  using a higher-order differential equation solver, hence we derive a new, faster
273
- technique - dpm-solver++ Inversion.
274
 
275
  </p>
276
 
 
263
  <p>
264
  Utilizing T2I models for editing real images is usually done by inverting the sampling
265
  process to identify a noisy xT that will be denoised to the input image x0.
266
+ We draw characteristics from <a href="https://inbarhub.github.io/DDPM_inversion/" target="_blank">edit friendly DDPM inversion</a> and propose an efficient
267
  inversion method that greatly reduces the required number
268
  of steps while maintaining no reconstruction error.
269
  DDPM can be viewed as a first-order
270
  SDE solver when formulating the reverse diffusion process as an SDE. This
271
  SDE can be solved more efficiently—in fewer steps—
272
  using a higher-order differential equation solver, hence we derive a new, faster
273
+ technique - <b>dpm-solver++ Inversion</b>.
274
 
275
  </p>
276