The resulting images are of poor quality

shaoyanpan / Synthetic-CT-generation-from-MRI-using-3D-transformer-based-denoising-diffusion-model

This is the repository for the paper published in Medical Physics: "Synthetic CT generation from MRI using 3D transformer-based denoising diffusion model".

https://aapm.onlinelibrary.wiley.com/doi/abs/10.1002/mp.16847

MIT License

28 stars 7 forks source link

The resulting images are of poor quality #10

Open dyf111 opened 4 days ago

dyf111 commented 4 days ago

Hello, teacher, the quality of the images generated by my training is not very good, what should I do

shaoyanpan commented 3 days ago

Sure. Firstly, can you tell me in this line, "predict_xstart=True"?

dyf111 commented 2 days ago

The predict_xstart parameter controls whether the model predicts the start image directly (x_0)...Should I change this parameter, teacher, because I noticed the note:# Don't toch these parameters, they are irrelant to the image synthesis

dyf111 commented 2 days ago

Teacher, I think I know where the mistake is, is the dimension of the data set I used not aligned? Should I follow img_size = (256,256,128) in the code? If so, how should I operate? Sorry, I am a beginner, just divide the data set and feed it directly to the network