How to adjust this result in coding？

[Explore008]

Hello, I used taichi model to drive people. The pixel of the image used is 256*256, and the video uses monotonous background, but the driving effect will produce artifacts, and the movement is not natural. I want to modify this part of the code, but I don't know which code to modify. Can you give me some advice? Thank you very much!

test4.mp4

result.mp4

[dazmashaly]

im sorry , how long did you train to get this result and did you train the avd also or just the model itself ?

[Explore008]

对不起，您训练了多长时间才能获得此结果，您是否也训练了 AVD 或只是训练了模型本身？
Hello! Your work is excellent! I didn't train, just tested your trained model with a photo, which will produce some shadows. I want to know whether taichi model can only be migrated in slow motion, and how to optimize this part of the shadow. Thank you!

[ykhasia]

buddy, this model should be "Artificial Intellectual disability", not "Artificial intelligence"