Hugging Face's logo

multimodalart
/
sdxl_perturbed_attention_guidance

Unconditional Image Generation
English
Diffusion Models
Stable Diffusion
Perturbed-Attention Guidance
PAG
Model card Files Community
2
multimodalart HF staff commited on
Commit
6a63c98
·
verified ·
1 Parent(s): e81c441

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -3
README.md CHANGED
@@ -10,8 +10,16 @@ tags:
10
  ---
11
 
12
  # Perturbed-Attention Guidance for SDXL
13
- ![image/jpeg](uncond_generation_pag.jpg)
14
- ![image/jpeg](cfgpag.jpg)
 
 
 
 
 
 
 
 
15
 
16
  [Project](https://ku-cvlab.github.io/Perturbed-Attention-Guidance/) / [arXiv](https://arxiv.org/abs/2403.17377) / [GitHub](https://github.com/KU-CVLAB/Perturbed-Attention-Guidance)
17
 
@@ -37,6 +45,7 @@ pipe = pipe.to(device)
37
  ```
38
 
39
  Unconditional sampling with PAG:
 
40
 
41
  ```py
42
  output = pipe(
@@ -49,7 +58,7 @@ output = pipe(
49
  ```
50
 
51
  Sampling with PAG and CFG:
52
-
53
  ```py
54
  output = pipe(
55
  "the spirit of a tamagotchi wandering in the city of Vienna",
 
10
  ---
11
 
12
  # Perturbed-Attention Guidance for SDXL
13
+
14
+ <div style="display:flex">
15
+ <video loop>
16
+ <source src="pag_sdxl.mp4" type="video/mp4">
17
+ </video>
18
+ <video loop>
19
+ <source src="pag_uncond.mp4" type="video/mp4">
20
+ </video>
21
+ </div>
22
+
23
 
24
  [Project](https://ku-cvlab.github.io/Perturbed-Attention-Guidance/) / [arXiv](https://arxiv.org/abs/2403.17377) / [GitHub](https://github.com/KU-CVLAB/Perturbed-Attention-Guidance)
25
 
 
45
  ```
46
 
47
  Unconditional sampling with PAG:
48
+ ![image/jpeg](uncond_generation_pag.jpg)
49
 
50
  ```py
51
  output = pipe(
 
58
  ```
59
 
60
  Sampling with PAG and CFG:
61
+ ![image/jpeg](cfgpag.jpg)
62
  ```py
63
  output = pipe(
64
  "the spirit of a tamagotchi wandering in the city of Vienna",