gouchaonijiao
gouchaonijiao
 Thanks for your reply！ But the problem is when change the shape of tgt_tensor(the parameter of "get_tgt_tensor.py"), it seems weird. I've also tried to extract openpose coordinate in size...
Thanks for your answer,I'll adjust the code into my desired size then. Thanks again!
Hello! Sorry to bother you again. I wonder know why you resize the image into (256,256) first when inference.? The generated image will undergo deformation or can you tell me...
I sincerely appreciate your patient answer and wish you complete more outstanding work in the future!
Since there are only 6 keypoints can be used, I'd like to make more keypoints to .npy file with your method. If you could reply to me,I would greatly appreciate...
Thanks for your reply! I will try it right away!
I want to express my gratitude to you again! The code you provided is very helpful to me ! Thanks very much
Thank you for your prompt reply! But I still can't understand very well. Frozen_out is not a function,so how these code " out = self.p_mean_variance( model, x_t, t, clip_denoised=clip_denoised, model_kwargs=model_kwargs...
Thans a lot for your detailed introduction,I sincerely appreciate your work and the dataset you made.By the way, how should I do the experiments if their code is not open-source...