syb7573330 / im2avatar Goto Github PK

Hi @syb7573330 I am trying to visualize the .h5 data, but I found the data had 4 dimensions. Could you tell me what's the meaning of each dimension?
Thanks a lot.

I got this error with tensorflow:latest, tensorflow:1.8.0-gpu-py3, 1.14.0-gpu-py3 and 1.14.0-gpu. I got them all via docker.

python train_shape.py --cat_id 02958343
Traceback (most recent call last):
  File "train_shape.py", line 51, in <module>
    LOG_FOUT.write(str(tf.flags._global_parser.parse_args())+'\n')
AttributeError: module 'tensorflow.python.platform.flags' has no attribute '_global_parser'

hi,
can anyone provide python code to plot the .h5 file. which is generated after running inference script.

Hi, I was wondering whether it was possible for you to upload the model parameters, for both shape and colors, for both of the mentioned datasets. If anyone else has completed their training, I would be glad if they can upload their model parameters here. Thanks!

I found that the voxel in output_shape_human is not 0 or 1，but range from [0,1]，I wonder what is the proper threshold to determine whether a voxel is filled.

Dear @syb7573330 I really like your work and wanted to cite your paper in my research. I want to ask how did you render the output in 3d model by .h5 files. Kindly guide me so that I can visualize them.

Best regards

Hi
I want to visualize the 3D outputs in output_shape and output_color folders.
How can I visualize them?

Could you please upload the subdataset to the Baidu Netdisk and share?We can not download it
in at home.Thank you! https://www.dropbox.com/s/s03fc1rx4ljkhix/ShapeNetCore_im2avatar.tar.gz

Can you please tell me the system requirements to train its model. Thanks in advance.

what is the version of TensorFlow？？