Text-to-Video
ChenyangSi nielsr HF staff commited on
Commit
aec7213
·
verified ·
1 Parent(s): c30c5bd

Add pipeline tag, link to project page (#1)

Browse files

- Add pipeline tag, link to project page (26dc5813eba1a75331e57ca46d3edf384b48e9e3)


Co-authored-by: Niels Rogge <[email protected]>

Files changed (1) hide show
  1. README.md +135 -3
README.md CHANGED
@@ -1,3 +1,135 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ pipeline_tag: text-to-video
4
+ ---
5
+
6
+ # RepVideo: Rethinking Cross-Layer Representation for Video Generation
7
+
8
+ <!-- <p align="center" width="100%">
9
+ <img src="ISEKAI_overview.png" width="80%" height="80%">
10
+ </p> -->
11
+
12
+ <div class="is-size-5 publication-authors", align="center">
13
+ <!-- Paper authors -->
14
+ <span class="author-block">
15
+ <a href="https://chenyangsi.top/" target="_blank">Chenyang Si</a><sup>1†</sup>,</span>
16
+ <span class="author-block">
17
+ <a href="https://scholar.google.com/citations?user=ORlELG8AAAAJ" target="_blank">Weichen Fan</a><sup>1†</sup>,</span>
18
+ <span class="author-block">
19
+ <a href="https://scholar.google.com/citations?user=FkkaUgwAAAAJ&hl=en" target="_blank">Zhengyao Lv</a><sup>2</sup>,</span>
20
+ <span class="author-block">
21
+ <a href="https://ziqihuangg.github.io/" target="_blank">Ziqi Huang</a><sup>1</sup>,</span>
22
+ <span class="author-block">
23
+ <a href="https://mmlab.siat.ac.cn/yuqiao" target="_blank">Yu Qiao</a><sup>2</sup>,</span>
24
+ <span class="author-block">
25
+ <a href="https://liuziwei7.github.io/" target="_blank">Ziwei Liu</a><sup>1✉</sup>
26
+ </span>
27
+ </div>
28
+ <div class="is-size-5 publication-authors", align="center">
29
+ <span class="author-block">S-Lab, Nanyang Technological University<sup>1</sup> &nbsp;&nbsp;&nbsp;&nbsp; Shanghai Artificial Intelligence Laboratory <sup>2</sup> </span>
30
+ <span class="eql-cntrb"><small><br><sup>†</sup>Equal contribution.&nbsp;&nbsp;&nbsp;&nbsp;<sup>✉</sup>Corresponding Author.</small></span>
31
+ </div>
32
+
33
+ </p>
34
+
35
+ <div align="center">
36
+ <a href="https://arxiv.org/pdf/2501.08994">Paper</a> |
37
+ <a href="https://vchitect.github.io/RepVid-Webpage/">Project Page</a>
38
+ </div>
39
+ <!-- <p align="center">
40
+ 👋 Join our <a href="https://github.com/Vchitect/RepVideo/tree/master/assets/channel/lark.jpeg" target="_blank">Lark</a> and <a href="https://discord.gg/aJAbn9sN" target="_blank">Discord</a>
41
+ </p> -->
42
+
43
+ ---
44
+
45
+ ![](https://img.shields.io/badge/RepVideo-v0.1-darkcyan)
46
+ [![Hits](https://hits.seeyoufarm.com/api/count/incr/badge.svg?url=https%3A%2F%2Fgithub.com%2FVchitect%2FRepVideo&count_bg=%23BDC4B7&title_bg=%2342C4A8&icon=octopusdeploy.svg&icon_color=%23E7E7E7&title=visitors&edge_flat=true)](https://hits.seeyoufarm.com)
47
+ [![Generic badge](https://img.shields.io/badge/Checkpoint-red.svg)](https://huggingface.co/Vchitect/RepVideo)
48
+ [![Hits](https://hits.seeyoufarm.com/api/count/incr/badge.svg?url=https%3A%2F%2Farxiv.org%2Fpdf%2F2501.08994&count_bg=%2379C83D&title_bg=%23555555&icon=&icon_color=%23E7E7E7&title=Paper&edge_flat=false)](https://hits.seeyoufarm.com)
49
+ [![Hits](https://hits.seeyoufarm.com/api/count/incr/badge.svg?url=https%3A%2F%2Fgithub.com%2FVchitect%2FRepVid-Webpage&count_bg=%23BE4C4C&title_bg=%235E5D64&icon=&icon_color=%23E7E7E7&title=Page&edge_flat=false)](https://hits.seeyoufarm.com)
50
+
51
+ ## :astonished: Gallery
52
+
53
+ <table class="center">
54
+ <tr>
55
+
56
+ <td><img src="assets/1.gif"> </td>
57
+ <td><img src="assets/2.gif"> </td>
58
+ <td><img src="assets/3.gif"> </td>
59
+ </tr>
60
+
61
+
62
+ <tr>
63
+ <td><img src="assets/4.gif"> </td>
64
+ <td><img src="assets/5.gif"> </td>
65
+ <td><img src="assets/6.gif"> </td>
66
+ </tr>
67
+
68
+ <tr>
69
+ <td><img src="assets/7.gif"> </td>
70
+ <td><img src="assets/8.gif"> </td>
71
+ <td><img src="assets/9.gif"> </td>
72
+ </tr>
73
+
74
+ <tr>
75
+ <td><img src="assets/10.gif"> </td>
76
+ <td><img src="assets/11.gif"> </td>
77
+ <td><img src="assets/12.gif"> </td>
78
+ </tr>
79
+
80
+ </table>
81
+
82
+ ## Installation
83
+
84
+ ### 1. Create a conda environment and download models
85
+
86
+
87
+ ```bash
88
+ conda create -n RepVid python==3.10
89
+ conda activate RepVid
90
+ pip install -r requirements.txt
91
+
92
+
93
+ mkdir ckpt
94
+ cd ckpt
95
+ mkdir t5-v1_1-xxl
96
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/config.json
97
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/model-00001-of-00002.safetensors
98
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/model-00002-of-00002.safetensors
99
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/model.safetensors.index.json
100
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/added_tokens.json
101
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/special_tokens_map.json
102
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/spiece.model
103
+ wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/tokenizer_config.json
104
+
105
+ cd ../
106
+ mkdir vae
107
+ wget https://cloud.tsinghua.edu.cn/f/fdba7608a49c463ba754/?dl=1
108
+ mv 'index.html?dl=1' vae.zip
109
+ unzip vae.zip
110
+ ```
111
+
112
+ ## Inference
113
+
114
+ ~~~bash
115
+ cd sat
116
+ bash run.sh
117
+ ~~~
118
+
119
+ ## BibTeX
120
+ ```
121
+ @article{si2025RepVideo,
122
+ title={RepVideo: Rethinking Cross-Layer Representation for Video Generation},
123
+ author={Si, Chenyang and Fan, Weichen and Lv, Zhengyao and Huang, Ziqi and Qiao, Yu and Liu, Ziwei},
124
+ journal={arXiv 2501.08994},
125
+ year={2025}
126
+ }
127
+ ```
128
+
129
+ ## 🔑 License
130
+
131
+ This code is licensed under Apache-2.0. The framework is fully open for academic research and also allows free commercial usage.
132
+
133
+ ## Disclaimer
134
+
135
+ We disclaim responsibility for user-generated content. The model was not trained to realistically represent people or events, so using it to generate such content is beyond the model's capabilities. It is prohibited for pornographic, violent and bloody content generation, and to generate content that is demeaning or harmful to people or their environment, culture, religion, etc. Users are solely liable for their actions. The project contributors are not legally affiliated with, nor accountable for users' behaviors. Use the generative model responsibly, adhering to ethical and legal standards.