PIDM
PIDM copied to clipboard
Person Image Synthesis via Denoising Diffusion Model (CVPR 2023)
If I want to use more key points like below, which file should I modify? ![image](https://github.com/ankanbhunia/PIDM/assets/98502973/943eb2a6-c890-485f-8f15-aa7199e43baa)
I haven't started working on this project yet, but it seems that the current pose is not controllable. That is, it seems infeasible if I want to generate other perspectives...
Hi authors, your work is impressive. Thanks for sharing the code base. However, I find the file "utils/metrics.py" is the evaluation code only for 256x176 images. And the FID calculated...
please share training weights on market1501-dataset
Hi thanks for the execellent work. May I know if you will release the trained model for the 512*352 resolution, so as to generate the 512*352 results ?
Hello! Your work is very impressive,but I wonder why the image and tgt_tensor are scaled to 256*256? And how I get other size of tgt_tensor like (256,512). Wish your reply!
Hello! Thanks for your amazing work! I want to know that how I make more "data/deepfashion_256x256/target_pose/*.npy" ?
when I train , why it becomes another people? ![model_408000](https://github.com/ankanbhunia/PIDM/assets/86581924/ba21fbe5-39a2-4c00-9aaf-9464427872bc)