how to keep the generated video the same size as the original image

fudan-generative-vision / hallo

Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animation

https://fudan-generative-vision.github.io/hallo/

MIT License

9.07k stars 1.22k forks source link

how to keep the generated video the same size as the original image #161

Open iraymondking opened 1 month ago

iraymondking commented 1 month ago

Is it possible to keep the generated video the same size as the original image? How to do it?

xumingw commented 1 month ago

We resize the image to 512x512 for model. Changing this hyperparameter may cause performance degrade. Some super resolution postprocessing may help if you really want to get more refined results.