Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,47 @@
|
|
1 |
-
---
|
2 |
-
license: mit
|
3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
---
|
4 |
+
# LEGO-Diffusion checkpoint repository
|
5 |
+
|
6 |
+
### For implementation details, please refer to our [Github Repo](https://github.com/JegZheng/LEGODiffusion)
|
7 |
+
|
8 |
+
|
9 |
+
#### ImageNet pretrained checkpoints
|
10 |
+
<table><tbody>
|
11 |
+
<!-- START TABLE -->
|
12 |
+
<!-- TABLE HEADER -->
|
13 |
+
<th align="left">model</th>
|
14 |
+
<th align="left">FID</th>
|
15 |
+
<th align="left">IS</th>
|
16 |
+
<!-- TABLE BODY -->
|
17 |
+
<tr>
|
18 |
+
<td align="left">LEGO-L-PG</td>
|
19 |
+
<td align="left">2.16</td>
|
20 |
+
<td align="left">68.7</td>
|
21 |
+
</tr>
|
22 |
+
<tr>
|
23 |
+
<td align="left">LEGO-L-PR</td>
|
24 |
+
<td align="left">2.29</td>
|
25 |
+
<td align="left">69.1</td>
|
26 |
+
</tr>
|
27 |
+
<tr>
|
28 |
+
<td align="left">LEGO-XL-U</td>
|
29 |
+
<td align="left">2.59</td>
|
30 |
+
<td align="left">338.08</td>
|
31 |
+
</tr>
|
32 |
+
</tbody></table>
|
33 |
+
|
34 |
+
|
35 |
+
### BibTex
|
36 |
+
Please cite our work if you find it is helpful. Thank you!
|
37 |
+
```
|
38 |
+
@inproceedings{
|
39 |
+
zheng2024learning,
|
40 |
+
title={Learning Stackable and Skippable {LEGO} Bricks for Efficient, Reconfigurable, and Variable-Resolution Diffusion Modeling},
|
41 |
+
author={Huangjie Zheng and Zhendong Wang and Jianbo Yuan and Guanghan Ning and Pengcheng He and Quanzeng You and Hongxia Yang and Mingyuan Zhou},
|
42 |
+
booktitle={The Twelfth International Conference on Learning Representations},
|
43 |
+
year={2024},
|
44 |
+
url={https://openreview.net/forum?id=qmXedvwrT1}
|
45 |
+
}
|
46 |
+
```
|
47 |
+
|