Abnormal transferring appearance results about idr HOT 7 OPEN

Hi .@SherlockSunset. I am also facing the same issue. and even I tried with the same appearance and same geometry id. still was not getting the right results

Same appearance and same geometry id? you mean you can't get right results even you just utilize the trained weights that the author provided for each case?

from idr.

@SherlockSunset Hi, could you tell me how you were able to get colored models?

from idr.

Hi .@SherlockSunset. I am also facing the same issue. and even I tried with the same appearance and same geometry id. still was not getting the right results

from idr.

@ahnaf1393 these are the rendered images not the colored models.

from idr.

@SherlockSunset I tried it for my own trained model using the evaluation/eval_disentanglement.py script. but not getting the right results.

from idr.

@SherlockSunset for the author provided model i am getting the following error "ValueError: num_samples should be a positive integer value, but got num_samples=0
"

from idr.

@SherlockSunset for the author provided model i am getting the following error "ValueError: num_samples should be a positive integer value, but got num_samples=0 "

Hi, maybe you can double check whether the dataset id you are using is correct or not.

from idr.