This Lora is used to convert pictures into 2D dynamic videos, similar to the effect of puppet animation. It is still in the experimental stage and has a slight gap from what I expected (the characters don't blink). For now, I'm releasing it for everyone to use. If many people use it, I will optimize the dataset and train another one.After testing, it was found that using the high noise model alone can produce results, and combining it with the low noise model can generate more details.Trigger Words：2D animation effects
If you forward this, please include the website address of this site and my name！！
You can use it for free online：https://www.runninghub.ai/ai-detail/1962768487540240386/?inviteCode=rh-v1167
❤️ Register and get 1000 points of computing power for free！ [You can generate over 100 images for free] Log in every day and get an additional 100 points of computing power！ ❤️
本lora用于将图片转换成2d动态视频，类似于木偶动画的效果。它尚且在试验阶段，跟我预想的有一点点差距（人物不会眨眼睛），目前先开放出来供大家使用，用的人多的话我会优化数据集后再次训练一个，经过测试，单独使用high noise模型就可以出效果，配合low noise模型可以生成更多的细节。触发词：2D animation effects
如果转发，请带上本站地址以及我的名字！！！
您可以免费在线使用：https://www.runninghub.ai/ai-detail/1962768487540240386/?inviteCode=rh-v1167
❤️ 注册即可免费获得 1000 点算力！[您可以免费生成超过 100 张图片] 每天登录还可额外获得 100 点算力！❤️
Description
FAQ
Comments (10)
It also works on real people's photos. Welcome everyone to share more pictures!
Appreciated the efforts. BTW have you ever experience unwanted moving mouth issue when using Wan2.2? I have been troubled by it for awhile....
yes, this happens to me very often. What I do usually to combat this is to add "talking, speaking" in the negative prompt. If you use lightx2v lightning lora, you also need to use WanVideoNAG node from KJNodes to make negative prompt taking effect.
This method doesn't work everytime, but will lower the chance of happening.
Please make a Version for WAN 2.2 5b
酷啊，可以自己做立绘了。
如果可以实现循环就好了,目前只能用工作流截尾帧到首帧,跑两段视频搞个循环,但是两段视频可能摇晃速度,风速,水流速度等有区别,转换的太突兀,要抽很多次卡才能挑一个不突兀的出来. 还有一点是镜头一直来回晃,如果镜头能稳定就好了
it intrsted. thx )
A major problem with Wan 2.2 is the lack of facial expression control.
This Lora is effective at making facial expressions more CALM.
I haven't tried using it to add effects. :D
最近初めてComfyUIでWan2.2をデビューしました。
動画生成のLoraはHighとLow2種類のLoraが必要と学んだのですがこのLoraはどうやって使用するのでしょうか？
使用しているワークフローはEasyWan22に入っていた00-I2v_ImageToVideoという物です。
It can be used simultaneously or separately under high_noise and low_noise.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.