Update README.md
Browse files
README.md
CHANGED
|
@@ -1,353 +1,356 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
language:
|
| 4 |
-
- en
|
| 5 |
-
- zh
|
| 6 |
-
pipeline_tag:
|
| 7 |
-
library_name:
|
| 8 |
-
tags:
|
| 9 |
-
- video
|
| 10 |
-
- video-generation
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
- [
|
| 28 |
-
- [
|
| 29 |
-
- [
|
| 30 |
-
- [
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
|
| 38 |
-
|
| 39 |
-
| Wan2.2-Fun-
|
| 40 |
-
| Wan2.2-Fun-
|
| 41 |
-
| Wan2.2-Fun-
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
| 54 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 58 |
-
|
| 59 |
-
<td>
|
| 60 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
<td>
|
| 71 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 72 |
-
</td>
|
| 73 |
-
|
| 74 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 75 |
-
|
| 76 |
-
|
| 77 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 78 |
-
</td>
|
| 79 |
-
|
| 80 |
-
|
| 81 |
-
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
<td>
|
| 91 |
-
|
| 92 |
-
</td>
|
| 93 |
-
<td>
|
| 94 |
-
|
| 95 |
-
</td>
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
<td>
|
| 101 |
-
<
|
| 102 |
-
</td>
|
| 103 |
-
|
| 104 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
|
| 108 |
-
|
| 109 |
-
|
| 110 |
-
|
| 111 |
-
|
| 112 |
-
|
| 113 |
-
|
| 114 |
-
|
| 115 |
-
|
| 116 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 117 |
-
|
| 118 |
-
|
| 119 |
-
|
| 120 |
-
|
| 121 |
-
<
|
| 122 |
-
|
| 123 |
-
|
| 124 |
-
|
| 125 |
-
|
| 126 |
-
<td>
|
| 127 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 128 |
-
</td>
|
| 129 |
-
|
| 130 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 131 |
-
|
| 132 |
-
|
| 133 |
-
|
| 134 |
-
|
| 135 |
-
|
| 136 |
-
<td>
|
| 137 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 138 |
-
</td>
|
| 139 |
-
|
| 140 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 141 |
-
|
| 142 |
-
|
| 143 |
-
|
| 144 |
-
|
| 145 |
-
|
| 146 |
-
|
| 147 |
-
|
| 148 |
-
|
| 149 |
-
|
| 150 |
-
|
| 151 |
-
|
| 152 |
-
<td>
|
| 153 |
-
Pan
|
| 154 |
-
</td>
|
| 155 |
-
|
| 156 |
-
Pan
|
| 157 |
-
|
| 158 |
-
|
| 159 |
-
|
| 160 |
-
</td>
|
| 161 |
-
|
| 162 |
-
|
| 163 |
-
|
| 164 |
-
|
| 165 |
-
<td>
|
| 166 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 167 |
-
</td>
|
| 168 |
-
|
| 169 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 170 |
-
|
| 171 |
-
|
| 172 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 173 |
-
</td>
|
| 174 |
-
|
| 175 |
-
|
| 176 |
-
|
| 177 |
-
|
| 178 |
-
<td>
|
| 179 |
-
Pan
|
| 180 |
-
</td>
|
| 181 |
-
|
| 182 |
-
Pan Up + Pan
|
| 183 |
-
|
| 184 |
-
<td>
|
| 185 |
-
|
| 186 |
-
</td>
|
| 187 |
-
|
| 188 |
-
|
| 189 |
-
|
| 190 |
-
|
| 191 |
-
<td>
|
| 192 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 193 |
-
</td>
|
| 194 |
-
|
| 195 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 196 |
-
|
| 197 |
-
|
| 198 |
-
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/
|
| 199 |
-
</td>
|
| 200 |
-
|
| 201 |
-
|
| 202 |
-
|
| 203 |
-
|
| 204 |
-
|
| 205 |
-
|
| 206 |
-
|
| 207 |
-
|
| 208 |
-
|
| 209 |
-
|
| 210 |
-
|
| 211 |
-
|
| 212 |
-
|
| 213 |
-
|
| 214 |
-
|
| 215 |
-
|
| 216 |
-
|
| 217 |
-
|
| 218 |
-
|
| 219 |
-
|
| 220 |
-
|
| 221 |
-
|
| 222 |
-
|
| 223 |
-
|
| 224 |
-
#
|
| 225 |
-
docker
|
| 226 |
-
|
| 227 |
-
#
|
| 228 |
-
|
| 229 |
-
|
| 230 |
-
#
|
| 231 |
-
|
| 232 |
-
|
| 233 |
-
#
|
| 234 |
-
|
| 235 |
-
|
| 236 |
-
|
| 237 |
-
|
| 238 |
-
|
| 239 |
-
|
| 240 |
-
#
|
| 241 |
-
|
| 242 |
-
#
|
| 243 |
-
# https://
|
| 244 |
-
|
| 245 |
-
#
|
| 246 |
-
# https://
|
| 247 |
-
|
| 248 |
-
|
| 249 |
-
|
| 250 |
-
|
| 251 |
-
|
| 252 |
-
|
| 253 |
-
|
| 254 |
-
|
| 255 |
-
|
| 256 |
-
|
| 257 |
-
-
|
| 258 |
-
-
|
| 259 |
-
-
|
| 260 |
-
|
| 261 |
-
|
| 262 |
-
-
|
| 263 |
-
|
| 264 |
-
|
| 265 |
-
-
|
| 266 |
-
-
|
| 267 |
-
-
|
| 268 |
-
|
| 269 |
-
|
| 270 |
-
|
| 271 |
-
|
| 272 |
-
|
| 273 |
-
|
| 274 |
-
|
| 275 |
-
|
| 276 |
-
|
| 277 |
-
|
| 278 |
-
|
| 279 |
-
|
| 280 |
-
|
| 281 |
-
|
| 282 |
-
│
|
| 283 |
-
│
|
| 284 |
-
|
| 285 |
-
|
| 286 |
-
|
| 287 |
-
```
|
| 288 |
-
|
| 289 |
-
|
| 290 |
-
|
| 291 |
-
|
| 292 |
-
|
| 293 |
-
│
|
| 294 |
-
├── 📂
|
| 295 |
-
│
|
| 296 |
-
|
| 297 |
-
|
| 298 |
-
|
| 299 |
-
|
| 300 |
-
|
| 301 |
-
|
| 302 |
-
|
| 303 |
-
|
| 304 |
-
|
| 305 |
-
|
| 306 |
-
-
|
| 307 |
-
|
| 308 |
-
|
| 309 |
-
|
| 310 |
-
|
| 311 |
-
|
| 312 |
-
|
| 313 |
-
|
| 314 |
-
####
|
| 315 |
-
|
| 316 |
-
|
| 317 |
-
|
| 318 |
-
|
| 319 |
-
|
| 320 |
-
-
|
| 321 |
-
- 使用examples/cogvideox_fun/
|
| 322 |
-
-
|
| 323 |
-
|
| 324 |
-
|
| 325 |
-
-
|
| 326 |
-
-
|
| 327 |
-
|
| 328 |
-
|
| 329 |
-
-
|
| 330 |
-
-
|
| 331 |
-
|
| 332 |
-
-
|
| 333 |
-
|
| 334 |
-
|
| 335 |
-
|
| 336 |
-
|
| 337 |
-
|
| 338 |
-
|
| 339 |
-
-
|
| 340 |
-
|
| 341 |
-
|
| 342 |
-
|
| 343 |
-
-
|
| 344 |
-
|
| 345 |
-
|
| 346 |
-
-
|
| 347 |
-
-
|
| 348 |
-
-
|
| 349 |
-
-
|
| 350 |
-
-
|
| 351 |
-
|
| 352 |
-
|
| 353 |
-
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
- zh
|
| 6 |
+
pipeline_tag: image-to-video
|
| 7 |
+
library_name: wan2.2
|
| 8 |
+
tags:
|
| 9 |
+
- video
|
| 10 |
+
- video-generation
|
| 11 |
+
- wan2.2
|
| 12 |
+
base_model:
|
| 13 |
+
- Wan-AI/Wan2.2-TI2V-5B
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# Wan-Fun
|
| 17 |
+
|
| 18 |
+
😊 Welcome!
|
| 19 |
+
|
| 20 |
+
[](https://huggingface.co/spaces/alibaba-pai/Wan2.1-Fun-1.3B-InP)
|
| 21 |
+
|
| 22 |
+
[](https://github.com/aigc-apps/VideoX-Fun)
|
| 23 |
+
|
| 24 |
+
[English](./README_en.md) | [简体中文](./README.md)
|
| 25 |
+
|
| 26 |
+
# 目录
|
| 27 |
+
- [目录](#目录)
|
| 28 |
+
- [模型地址](#模型地址)
|
| 29 |
+
- [视频作品](#视频作品)
|
| 30 |
+
- [快速启动](#快速启动)
|
| 31 |
+
- [如何使用](#如何使用)
|
| 32 |
+
- [参考文献](#参考文献)
|
| 33 |
+
- [许可证](#许可证)
|
| 34 |
+
|
| 35 |
+
# 模型地址
|
| 36 |
+
|
| 37 |
+
| 名称 | 存储空间 | Hugging Face | Model Scope | 描述 |
|
| 38 |
+
|--|--|--|--|--|
|
| 39 |
+
| Wan2.2-Fun-A14B-InP | 64.0 GB | [🤗Link](https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-InP) | [😄Link](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP) | Wan2.2-Fun-14B文图生视频权重，以多分辨率训练，支持首尾图预测。 |
|
| 40 |
+
| Wan2.2-Fun-A14B-Control | 64.0 GB | [🤗Link](https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-Control) | [😄Link](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control)| Wan2.2-Fun-14B视频控制权重，支持不同的控制条件，如Canny、Depth、Pose、MLSD等，同时支持使用轨迹控制。支持多分辨率（512，768，1024）的视频预测，，以81帧、每秒16帧进行训练，支持多语言预测 |
|
| 41 |
+
| Wan2.2-Fun-A14B-Control-Camera | 64.0 GB | [🤗Link](https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-Control-Camera) | [😄Link](https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-Control-Camera)| Wan2.2-Fun-14B相机镜头控制权重。支持多分辨率（512，768，1024）的视频预测，，以81帧、每秒16帧进行训练，支持多语言预测 |
|
| 42 |
+
| Wan2.2-Fun-5B-InP | 23.0 GB | [🤗Link](https://huggingface.co/alibaba-pai/Wan2.2-Fun-5B-InP) | [😄Link](https://modelscope.cn/models/PAI/Wan2.2-Fun-5B-InP) | Wan2.2-Fun-5B文图生视频权重，以121帧、每秒24帧进行训练支持首尾图预测。 |
|
| 43 |
+
| Wan2.2-Fun-5B-Control | 23.0 GB | [🤗Link](https://huggingface.co/alibaba-pai/Wan2.2-Fun-5B-Control) | [😄Link](https://modelscope.cn/models/PAI/Wan2.2-Fun-5B-Control)| Wan2.2-Fun-5B视频控制权重，支持不同的控制条件，如Canny、Depth、Pose、MLSD等，同时支持使用轨迹控制。以121帧、每秒24帧进行训练，支持多语言��测 |
|
| 44 |
+
| Wan2.2-Fun-5B-Control-Camera | 23.0 GB | [🤗Link](https://huggingface.co/alibaba-pai/Wan2.2-Fun-5B-Control-Camera) | [😄Link](https://modelscope.cn/models/PAI/Wan2.2-Fun-5B-Control-Camera)| Wan2.2-Fun-5B相机镜头控制权重。以121帧、每秒24帧进行训练，支持多语言预测 |
|
| 45 |
+
|
| 46 |
+
|
| 47 |
+
# 视频作品
|
| 48 |
+
|
| 49 |
+
### Wan2.2-Fun-A14B-InP
|
| 50 |
+
|
| 51 |
+
<table border="0" style="width: 100%; text-align: left; margin-top: 20px;">
|
| 52 |
+
<tr>
|
| 53 |
+
<td>
|
| 54 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_1.mp4" width="100%" controls autoplay loop></video>
|
| 55 |
+
</td>
|
| 56 |
+
<td>
|
| 57 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_2.mp4" width="100%" controls autoplay loop></video>
|
| 58 |
+
</td>
|
| 59 |
+
<td>
|
| 60 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_3.mp4" width="100%" controls autoplay loop></video>
|
| 61 |
+
</td>
|
| 62 |
+
<td>
|
| 63 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_4.mp4" width="100%" controls autoplay loop></video>
|
| 64 |
+
</td>
|
| 65 |
+
</tr>
|
| 66 |
+
</table>
|
| 67 |
+
|
| 68 |
+
<table border="0" style="width: 100%; text-align: left; margin-top: 20px;">
|
| 69 |
+
<tr>
|
| 70 |
+
<td>
|
| 71 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_5.mp4" width="100%" controls autoplay loop></video>
|
| 72 |
+
</td>
|
| 73 |
+
<td>
|
| 74 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_6.mp4" width="100%" controls autoplay loop></video>
|
| 75 |
+
</td>
|
| 76 |
+
<td>
|
| 77 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_7.mp4" width="100%" controls autoplay loop></video>
|
| 78 |
+
</td>
|
| 79 |
+
<td>
|
| 80 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/inp_8.mp4" width="100%" controls autoplay loop></video>
|
| 81 |
+
</td>
|
| 82 |
+
</tr>
|
| 83 |
+
</table>
|
| 84 |
+
|
| 85 |
+
### Wan2.2-Fun-A14B-Control
|
| 86 |
+
|
| 87 |
+
Generic Control Video + Reference Image:
|
| 88 |
+
<table border="0" style="width: 100%; text-align: left; margin-top: 20px;">
|
| 89 |
+
<tr>
|
| 90 |
+
<td>
|
| 91 |
+
Reference Image
|
| 92 |
+
</td>
|
| 93 |
+
<td>
|
| 94 |
+
Control Video
|
| 95 |
+
</td>
|
| 96 |
+
<td>
|
| 97 |
+
Wan2.2-Fun-14B-Control
|
| 98 |
+
</td>
|
| 99 |
+
<tr>
|
| 100 |
+
<td>
|
| 101 |
+
<image src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/8.png" width="100%" controls autoplay loop></image>
|
| 102 |
+
</td>
|
| 103 |
+
<td>
|
| 104 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/pose.mp4" width="100%" controls autoplay loop></video>
|
| 105 |
+
</td>
|
| 106 |
+
<td>
|
| 107 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/14b_ref.mp4" width="100%" controls autoplay loop></video>
|
| 108 |
+
</td>
|
| 109 |
+
<tr>
|
| 110 |
+
</table>
|
| 111 |
+
|
| 112 |
+
Generic Control Video (Canny, Pose, Depth, etc.) and Trajectory Control:
|
| 113 |
+
<table border="0" style="width: 100%; text-align: left; margin-top: 20px;">
|
| 114 |
+
<tr>
|
| 115 |
+
<td>
|
| 116 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/guiji.mp4" width="100%" controls autoplay loop></video>
|
| 117 |
+
</td>
|
| 118 |
+
<td>
|
| 119 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/guiji_out.mp4" width="100%" controls autoplay loop></video>
|
| 120 |
+
</td>
|
| 121 |
+
<tr>
|
| 122 |
+
</table>
|
| 123 |
+
|
| 124 |
+
<table border="0" style="width: 100%; text-align: left; margin-top: 20px;">
|
| 125 |
+
<tr>
|
| 126 |
+
<td>
|
| 127 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/pose.mp4" width="100%" controls autoplay loop></video>
|
| 128 |
+
</td>
|
| 129 |
+
<td>
|
| 130 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/canny.mp4" width="100%" controls autoplay loop></video>
|
| 131 |
+
</td>
|
| 132 |
+
<td>
|
| 133 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/depth.mp4" width="100%" controls autoplay loop></video>
|
| 134 |
+
</td>
|
| 135 |
+
<tr>
|
| 136 |
+
<td>
|
| 137 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/pose_out.mp4" width="100%" controls autoplay loop></video>
|
| 138 |
+
</td>
|
| 139 |
+
<td>
|
| 140 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/canny_out.mp4" width="100%" controls autoplay loop></video>
|
| 141 |
+
</td>
|
| 142 |
+
<td>
|
| 143 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/depth_out.mp4" width="100%" controls autoplay loop></video>
|
| 144 |
+
</td>
|
| 145 |
+
</tr>
|
| 146 |
+
</table>
|
| 147 |
+
|
| 148 |
+
### Wan2.2-Fun-A14B-Control-Camera
|
| 149 |
+
|
| 150 |
+
<table border="0" style="width: 100%; text-align: left; margin-top: 20px;">
|
| 151 |
+
<tr>
|
| 152 |
+
<td>
|
| 153 |
+
Pan Up
|
| 154 |
+
</td>
|
| 155 |
+
<td>
|
| 156 |
+
Pan Left
|
| 157 |
+
</td>
|
| 158 |
+
<td>
|
| 159 |
+
Pan Right
|
| 160 |
+
</td>
|
| 161 |
+
<td>
|
| 162 |
+
Zoom In
|
| 163 |
+
</td>
|
| 164 |
+
<tr>
|
| 165 |
+
<td>
|
| 166 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Pan_Up.mp4" width="100%" controls autoplay loop></video>
|
| 167 |
+
</td>
|
| 168 |
+
<td>
|
| 169 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Pan_Left.mp4" width="100%" controls autoplay loop></video>
|
| 170 |
+
</td>
|
| 171 |
+
<td>
|
| 172 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Pan_Right.mp4" width="100%" controls autoplay loop></video>
|
| 173 |
+
</td>
|
| 174 |
+
<td>
|
| 175 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Zoom_In.mp4" width="100%" controls autoplay loop></video>
|
| 176 |
+
</td>
|
| 177 |
+
<tr>
|
| 178 |
+
<td>
|
| 179 |
+
Pan Down
|
| 180 |
+
</td>
|
| 181 |
+
<td>
|
| 182 |
+
Pan Up + Pan Left
|
| 183 |
+
</td>
|
| 184 |
+
<td>
|
| 185 |
+
Pan Up + Pan Right
|
| 186 |
+
</td>
|
| 187 |
+
<td>
|
| 188 |
+
Zoom Out
|
| 189 |
+
</td>
|
| 190 |
+
<tr>
|
| 191 |
+
<td>
|
| 192 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Pan_Down.mp4" width="100%" controls autoplay loop></video>
|
| 193 |
+
</td>
|
| 194 |
+
<td>
|
| 195 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Pan_UL.mp4" width="100%" controls autoplay loop></video>
|
| 196 |
+
</td>
|
| 197 |
+
<td>
|
| 198 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Pan_UR.mp4" width="100%" controls autoplay loop></video>
|
| 199 |
+
</td>
|
| 200 |
+
<td>
|
| 201 |
+
<video src="https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/wan_fun/asset_Wan2_2/v1.0/Zoom_Out.mp4" width="100%" controls autoplay loop></video>
|
| 202 |
+
</td>
|
| 203 |
+
</tr>
|
| 204 |
+
</table>
|
| 205 |
+
|
| 206 |
+
|
| 207 |
+
# 快速启动
|
| 208 |
+
### 1. 云使用: AliyunDSW/Docker
|
| 209 |
+
#### a. 通过阿里云 DSW
|
| 210 |
+
DSW 有免费 GPU 时间，用户可申请一次，申请后3个月内有效。
|
| 211 |
+
|
| 212 |
+
阿里云在[Freetier](https://free.aliyun.com/?product=9602825&crowd=enterprise&spm=5176.28055625.J_5831864660.1.e939154aRgha4e&scm=20140722.M_9974135.P_110.MO_1806-ID_9974135-MID_9974135-CID_30683-ST_8512-V_1)提供免费GPU时间，获取并在阿里云PAI-DSW中使用，5分钟内即可启动CogVideoX-Fun。
|
| 213 |
+
|
| 214 |
+
[](https://gallery.pai-ml.com/#/preview/deepLearning/cv/cogvideox_fun)
|
| 215 |
+
|
| 216 |
+
#### b. 通过ComfyUI
|
| 217 |
+
我们的ComfyUI界面如下，具体查看[ComfyUI README](comfyui/README.md)。
|
| 218 |
+
|
| 219 |
+
|
| 220 |
+
#### c. 通过docker
|
| 221 |
+
使用docker的情况下，请保证机器中已经正确安装显卡驱动与CUDA环境，然后以此执行以下命令：
|
| 222 |
+
|
| 223 |
+
```
|
| 224 |
+
# pull image
|
| 225 |
+
docker pull mybigpai-public-registry.cn-beijing.cr.aliyuncs.com/easycv/torch_cuda:cogvideox_fun
|
| 226 |
+
|
| 227 |
+
# enter image
|
| 228 |
+
docker run -it -p 7860:7860 --network host --gpus all --security-opt seccomp:unconfined --shm-size 200g mybigpai-public-registry.cn-beijing.cr.aliyuncs.com/easycv/torch_cuda:cogvideox_fun
|
| 229 |
+
|
| 230 |
+
# clone code
|
| 231 |
+
git clone https://github.com/aigc-apps/VideoX-Fun.git
|
| 232 |
+
|
| 233 |
+
# enter VideoX-Fun's dir
|
| 234 |
+
cd VideoX-Fun
|
| 235 |
+
|
| 236 |
+
# download weights
|
| 237 |
+
mkdir models/Diffusion_Transformer
|
| 238 |
+
mkdir models/Personalized_Model
|
| 239 |
+
|
| 240 |
+
# Please use the hugginface link or modelscope link to download the model.
|
| 241 |
+
# CogVideoX-Fun
|
| 242 |
+
# https://huggingface.co/alibaba-pai/CogVideoX-Fun-V1.1-5b-InP
|
| 243 |
+
# https://modelscope.cn/models/PAI/CogVideoX-Fun-V1.1-5b-InP
|
| 244 |
+
|
| 245 |
+
# Wan
|
| 246 |
+
# https://huggingface.co/alibaba-pai/Wan2.1-Fun-V1.1-14B-InP
|
| 247 |
+
# https://modelscope.cn/models/PAI/Wan2.1-Fun-V1.1-14B-InP
|
| 248 |
+
# https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-InP
|
| 249 |
+
# https://modelscope.cn/models/PAI/Wan2.2-Fun-A14B-InP
|
| 250 |
+
```
|
| 251 |
+
|
| 252 |
+
### 2. 本地安装: 环境检查/下载/安装
|
| 253 |
+
#### a. 环境检查
|
| 254 |
+
我们已验证该库可在以下环境中执行：
|
| 255 |
+
|
| 256 |
+
Windows 的详细信息：
|
| 257 |
+
- 操作系统 Windows 10
|
| 258 |
+
- python: python3.10 & python3.11
|
| 259 |
+
- pytorch: torch2.2.0
|
| 260 |
+
- CUDA: 11.8 & 12.1
|
| 261 |
+
- CUDNN: 8+
|
| 262 |
+
- GPU： Nvidia-3060 12G & Nvidia-3090 24G
|
| 263 |
+
|
| 264 |
+
Linux 的详细信息：
|
| 265 |
+
- 操作系统 Ubuntu 20.04, CentOS
|
| 266 |
+
- python: python3.10 & python3.11
|
| 267 |
+
- pytorch: torch2.2.0
|
| 268 |
+
- CUDA: 11.8 & 12.1
|
| 269 |
+
- CUDNN: 8+
|
| 270 |
+
- GPU：Nvidia-V100 16G & Nvidia-A10 24G & Nvidia-A100 40G & Nvidia-A100 80G
|
| 271 |
+
|
| 272 |
+
我们需要大约 60GB 的可用磁盘空间，请检查！
|
| 273 |
+
|
| 274 |
+
#### b. 权重放置
|
| 275 |
+
我们最好将[权重](#model-zoo)按照指定路径进行放置：
|
| 276 |
+
|
| 277 |
+
**通过comfyui**：
|
| 278 |
+
将模型放入Comfyui的权重文件夹`ComfyUI/models/Fun_Models/`：
|
| 279 |
+
```
|
| 280 |
+
📦 ComfyUI/
|
| 281 |
+
├── 📂 models/
|
| 282 |
+
│ └── 📂 Fun_Models/
|
| 283 |
+
│ ├── 📂 CogVideoX-Fun-V1.1-2b-InP/
|
| 284 |
+
│ ├── 📂 CogVideoX-Fun-V1.1-5b-InP/
|
| 285 |
+
│ ├── 📂 Wan2.1-Fun-V1.1-14B-InP
|
| 286 |
+
│ └── 📂 Wan2.1-Fun-V1.1-1.3B-InP/
|
| 287 |
+
```
|
| 288 |
+
|
| 289 |
+
**运行自身的python文件或ui界面**:
|
| 290 |
+
```
|
| 291 |
+
📦 models/
|
| 292 |
+
├── 📂 Diffusion_Transformer/
|
| 293 |
+
│ ├── 📂 CogVideoX-Fun-V1.1-2b-InP/
|
| 294 |
+
│ ├── 📂 CogVideoX-Fun-V1.1-5b-InP/
|
| 295 |
+
│ ├── 📂 Wan2.1-Fun-V1.1-14B-InP
|
| 296 |
+
│ └── 📂 Wan2.1-Fun-V1.1-1.3B-InP/
|
| 297 |
+
├── 📂 Personalized_Model/
|
| 298 |
+
│ └── your trained trainformer model / your trained lora model (for UI load)
|
| 299 |
+
```
|
| 300 |
+
|
| 301 |
+
# 如何使用
|
| 302 |
+
|
| 303 |
+
<h3 id="video-gen">1. 生成 </h3>
|
| 304 |
+
|
| 305 |
+
#### a、显存节省方案
|
| 306 |
+
由于Wan2.2的参数非常大，我们需要考虑显存节省方案，以节省显存适应消费级显卡。我们给每个预测文件都提供了GPU_memory_mode，可以在model_cpu_offload，model_cpu_offload_and_qfloat8，sequential_cpu_offload中进行选择。该方案同样适用于CogVideoX-Fun的生成。
|
| 307 |
+
|
| 308 |
+
- model_cpu_offload代表整个模型在使用后会进入cpu，可以节省部分显存。
|
| 309 |
+
- model_cpu_offload_and_qfloat8代表整个模型在使用后会进入cpu，并且对transformer模型进行了float8的量化，可以节省更多的显存。
|
| 310 |
+
- sequential_cpu_offload代表模型的每一层在使用后会进入cpu，速度较慢，节省大量显存。
|
| 311 |
+
|
| 312 |
+
qfloat8会部分降低模型的性能，但可以节省更多的显存。如果显存足够，推荐使用model_cpu_offload。
|
| 313 |
+
|
| 314 |
+
#### b、通过comfyui
|
| 315 |
+
具体查看[ComfyUI README](https://github.com/aigc-apps/VideoX-Fun/tree/main/comfyui)。
|
| 316 |
+
|
| 317 |
+
#### c、运行python文件
|
| 318 |
+
- 步骤1：下载对应[权重](#model-zoo)放入models文件夹。
|
| 319 |
+
- 步骤2：根据不同的权重与预测目标使用不同的文件进行预测。当前该库支持CogVideoX-Fun、Wan2.1、Wan2.1-Fun、Wan2.2，在examples文件夹下用文件夹名以区分，不同模型支持的功能不同，请视具体情况予以区分。以CogVideoX-Fun为例。
|
| 320 |
+
- 文生视频：
|
| 321 |
+
- 使用examples/cogvideox_fun/predict_t2v.py文件中修改prompt、neg_prompt、guidance_scale和seed。
|
| 322 |
+
- 而后运行examples/cogvideox_fun/predict_t2v.py文件，等待生成结果，结果保存在samples/cogvideox-fun-videos文件夹中。
|
| 323 |
+
- 图生视频：
|
| 324 |
+
- 使用examples/cogvideox_fun/predict_i2v.py文件中修改validation_image_start、validation_image_end、prompt、neg_prompt、guidance_scale和seed。
|
| 325 |
+
- validation_image_start是视频的开始图片，validation_image_end是视频的结尾图片。
|
| 326 |
+
- 而后运行examples/cogvideox_fun/predict_i2v.py文件，等待生成结果，结果保存在samples/cogvideox-fun-videos_i2v文件夹中。
|
| 327 |
+
- 视频生视频：
|
| 328 |
+
- 使用examples/cogvideox_fun/predict_v2v.py文件中修改validation_video、validation_image_end、prompt、neg_prompt、guidance_scale和seed。
|
| 329 |
+
- validation_video是视频生视频的参考视频。您可以使用以下视频运行演示：[演示视频](https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/cogvideox_fun/asset/v1/play_guitar.mp4)
|
| 330 |
+
- 而后运行examples/cogvideox_fun/predict_v2v.py文件，等待生成结果，结果保存在samples/cogvideox-fun-videos_v2v文件夹中。
|
| 331 |
+
- 普通控制生视频（Canny、Pose、Depth等）：
|
| 332 |
+
- 使用examples/cogvideox_fun/predict_v2v_control.py文件中修改control_video、validation_image_end、prompt、neg_prompt、guidance_scale和seed。
|
| 333 |
+
- control_video是控制生视频的控制视频，是使用Canny、Pose、Depth等算子提取后的视频。您可以使用以下视频运行演示：[演示视频](https://pai-aigc-photog.oss-cn-hangzhou.aliyuncs.com/cogvideox_fun/asset/v1.1/pose.mp4)
|
| 334 |
+
- 而后运行examples/cogvideox_fun/predict_v2v_control.py文件，等待生成结果，结果保存在samples/cogvideox-fun-videos_v2v_control文件夹中。
|
| 335 |
+
- 步骤3：如果想结合自己训练的其他backbone与Lora，则看情况修改examples/{model_name}/predict_t2v.py中的examples/{model_name}/predict_i2v.py和lora_path。
|
| 336 |
+
|
| 337 |
+
#### d、通过ui界面
|
| 338 |
+
|
| 339 |
+
webui支持文生视频、图生视频、视频生视频和普通控制生视频（Canny、Pose、Depth等）。在examples文件夹下用文件夹名以区分，不同模型支持的功能不同，请视具体情况予以区分。以CogVideoX-Fun为例。
|
| 340 |
+
|
| 341 |
+
- 步骤1：下载对应[权重](#model-zoo)放入models文件夹。
|
| 342 |
+
- 步骤2：运行examples/cogvideox_fun/app.py文件，进入gradio页面。
|
| 343 |
+
- 步骤3：根据页面选择生成模型，填入prompt、neg_prompt、guidance_scale和seed等，点击生成，等待生成结果，结果保存在sample文件夹中。
|
| 344 |
+
|
| 345 |
+
# 参考文献
|
| 346 |
+
- CogVideo: https://github.com/THUDM/CogVideo/
|
| 347 |
+
- EasyAnimate: https://github.com/aigc-apps/EasyAnimate
|
| 348 |
+
- Wan2.1: https://github.com/Wan-Video/Wan2.1/
|
| 349 |
+
- Wan2.1: https://github.com/Wan-Video/Wan2.2/
|
| 350 |
+
- ComfyUI-KJNodes: https://github.com/kijai/ComfyUI-KJNodes
|
| 351 |
+
- ComfyUI-EasyAnimateWrapper: https://github.com/kijai/ComfyUI-EasyAnimateWrapper
|
| 352 |
+
- ComfyUI-CameraCtrl-Wrapper: https://github.com/chaojie/ComfyUI-CameraCtrl-Wrapper
|
| 353 |
+
- CameraCtrl: https://github.com/hehao13/CameraCtrl
|
| 354 |
+
|
| 355 |
+
# 许可证
|
| 356 |
+
本项目采用 [Apache License (Version 2.0)](https://github.com/modelscope/modelscope/blob/master/LICENSE).
|