Update README.md
Browse files
README.md
CHANGED
|
@@ -11,7 +11,7 @@ This is the Official Repository of "[Learning Few-Step Diffusion Models by Traje
|
|
| 11 |
|
| 12 |
|
| 13 |
## User Study Time!
|
| 14 |
-
. Some images are generated by **TDM (4 NFE)**, distilling from Pixart-α in a data-free way with merely 500 training iterations and 2 A800 hours.
|
| 16 |
|
| 17 |
<details>
|
|
@@ -27,8 +27,8 @@ Which one do you think is better? Some images are generated by Pixart-α (50 NFE
|
|
| 27 |
Our proposed TDM can be easily extended to text-to-video.
|
| 28 |
|
| 29 |
<p align="center">
|
| 30 |
-
<img src="
|
| 31 |
-
<img src="
|
| 32 |
</p>
|
| 33 |
|
| 34 |
The video on the left was generated by CogVideoX-2B (100 NFE). In the same amount of time, **TDM (4NFE)** can generate 25 videos, as shown on the right, achieving an impressive **25 times speedup without performance degradation**. (Note: The noise in the GIF is due to compression.)
|
|
|
|
| 11 |
|
| 12 |
|
| 13 |
## User Study Time!
|
| 14 |
+
|
| 15 |
Which one do you think is better? Some images are generated by Pixart-α (50 NFE). Some images are generated by **TDM (4 NFE)**, distilling from Pixart-α in a data-free way with merely 500 training iterations and 2 A800 hours.
|
| 16 |
|
| 17 |
<details>
|
|
|
|
| 27 |
Our proposed TDM can be easily extended to text-to-video.
|
| 28 |
|
| 29 |
<p align="center">
|
| 30 |
+
<img src="teacher.gif" alt="Teacher" width="45%">
|
| 31 |
+
<img src="student.gif" alt="Student" width="45%">
|
| 32 |
</p>
|
| 33 |
|
| 34 |
The video on the left was generated by CogVideoX-2B (100 NFE). In the same amount of time, **TDM (4NFE)** can generate 25 videos, as shown on the right, achieving an impressive **25 times speedup without performance degradation**. (Note: The noise in the GIF is due to compression.)
|