gafniguy / 4D-Facial-Avatars

Dynamic Neural Radiance Fields for Monocular 4D Facial Avater Reconstruction
670 stars 67 forks source link

How to achieve the same effect of facial replay as the paper, and whether it can achieve real-time? #29

Closed Chopin68 closed 1 year ago

Chopin68 commented 2 years ago

image When I follow the steps to generate the image of this step, what should I do next,Hope to get guidance!

gafniguy commented 2 years ago

Line 420 in the eval script is causing the pose and expression to be fixed. Set it to None

Chopin68 commented 2 years ago

eval 脚本中的第 420 行导致姿势和表情被固定。将其设置为无

Thank you for your reply!! I reported an error after running the following code

python real_to_nerf.py --source K:/GAN/4D-Facial-Avatars-main/nerface_code/nerf-pytorch/real_data/person_1 --target K:/GAN/4D-Facial-Avatars-main/nerface_code/nerf-pytorch/renders/person_1_rendered_frames

Traceback (most recent call last): File "K:\GAN\4D-Facial-Avatars-main\nerface_code\nerf-pytorch\real_to_nerf.py", line 7, in proc1 = subprocess.Popen(['scontrol', 'show', 'job', os.environ['SLURM_JOBID'], '-d'], stdout=subprocess.PIPE) File "C:\Users\CHOPY\AppData\Local\Programs\Python\Python39\lib\os.py", line 679, in getitem raise KeyError(key) from None KeyError: 'SLURM_JOBID'