ankanbhunia / PIDM

Person Image Synthesis via Denoising Diffusion Model (CVPR 2023)

https://ankanbhunia.github.io/PIDM

ankanbhunia/PIDM Issues

About pose (keypoint)
Updated 3 months ago
uncontrol pose ?
Updated 3 months ago
What are the versions of tensorfn and pydantic
Updated 3 months ago2
it seems import error with new version of pydantic library
Closed 4 months ago
Quantitative Problem
Closed a year ago26
.
Closed 4 months ago
results and evaluation for 512x352 images
Updated 4 months ago5
Market1501 Weights
Updated 4 months ago2
Where to get the model for 512*352 resolution ?
Updated 4 months ago1
about the size of tgt_pose when inference
Updated 5 months ago7
About the tgt_pose when inference
Updated 5 months ago8
predict_pose can not get good result when if resize target_pose npy to 1024x1024
Updated 6 months ago
when I train , why it becomes another people?
Updated 6 months ago2
Using the pre-trained model for market-1501 dataset
Updated 7 months ago5
About market-1501
Updated 7 months ago
when p_mean_variance, if x_cond is None: how can this code"if cond is None: x_cond = (cond_mask.view(-1,1,1,1)*x_cond)" run?
Closed a year ago1
did you try to train the model in fp16?
Updated 7 months ago
How long does it take to train 300 epoech
Updated 8 months ago
IndexError: list index out of range
Closed 8 months ago
about Train.py
Updated 8 months ago3
what is the password of img_highres.zip ?
Closed 9 months ago3
0
Closed 8 months ago
0
Closed 8 months ago
About the color difference between the generated image and ground truth
Updated 9 months ago1
how to generate the train_pairs.txt for my dataset ?
Updated 9 months ago1
keypoints to .npy
Updated 9 months ago7
CUDA out of memory !
Updated 9 months ago1
The possible contradiction for disentangled cfg between the paper and train code
Updated 9 months ago1
About Market-1501 Experiments
Closed 9 months ago1
Updated it to PyTorch2.0 and replaced Attention with PyTorch's Multihead attention
Updated 9 months ago1
About keypoints estimation
Closed a year ago4
how to generate the reference_pose_0.npy from image?
Updated 9 months ago
About GPUs
Closed a year ago2
how to train and test with a 11 G gpu ?
Updated 10 months ago
The Question about Figure 3 in Paper?
Closed 10 months ago2
About the implementation on multi-scale condition.
Updated 10 months ago1
hi, can you provide the model checkpoint of 512x352 ?
Updated a year ago2
Feature request: Run this in image to image style for generation.
Updated a year ago
SSIM, FID and PSNR
Closed a year ago2
The result of running the model is not so perfect
Closed a year ago1
About cond_scale
Updated a year ago1
About the training images
Closed a year ago7
How to start with the code
Closed a year ago1
About the ddim_steps function in deffusion.py
Closed a year ago2
custom dataset training
Closed a year ago5
Error while doing inference on resized PNG image
Closed a year ago2
Generalization of anime characters
Closed a year ago1
is there any ablation study of TDB reference feature size?
Updated a year ago1
About the model structure
Updated a year ago1
can you teach me how the"frozen_out" work? thanks!
Updated a year ago2