mohamedhassanmus / posa Goto Github PK

Thanks for your amazing work!
I notice that in the paper a UE scene is tested using the proposed method, I wonder how to get the corresponding point cloud and its semantic label?

Hi

It is an excellent work for modeling human scene interaction and I have a question about the affordance experiment.
As the following code:

Waiting for your reply.

Best wishes

Hi, thank you for the amazing work!
I followed the instructions and placed the free rp models (obj) in the rp_clothed_meshes folder. Under the same folder there is a tex folder with the textures. But it seems that your code is not picking up the textures. What tools or code did you use to generate the clothed human renderings as in your paper?

Also, the output of the rp model seems to be under the floor, is that a failure case of this method or is there a bug in the code?

Hi, thank you for the great work!
I wonder how you produced files (e.g. A_0.npz) in mesh_ds folder? I would like to produce such files for SMPL and SMIL models, and I am wondering how I could do that.
Thanks!

Hi ! Thanks for your great work.

I'm trying to make the affordance code work with my custom scenes. I tried replicating the .json file that describe the scene mesh but i don't quite understand how the min and max vectors are calculated.

Thanks in advance

Traceback (most recent call last):
  File "src/affordance.py", line 105, in <module>
    vertices_org, vertices_can, faces_arr, body_model, R_can, pelvis, torch_param, vertices_clothed = data_utils.pkl_to_canonical(
  File "<path>/POSA/src/data_utils.py", line 83, in pkl_to_canonical
    body_model.reset_params(**torch_param)
  File "<path>/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 28, in decorate_context
    return func(*args, **kwargs)
  File “<path>/lib/python3.8/site-packages/smplx-0.1.28-py3.8.egg/smplx/body_models.py", line 289, in reset_params
RuntimeError: The expanded size of the tensor (16) must match the existing size (10) at non-singleton dimension 1.  Target sizes: [1, 16].  Tensor sizes: [10]

The above Runtime Error is coming when running the sample commands like:

python src/affordance.py --config cfg_files/contact_semantics.yaml --checkpoint_path $POSA_dir/trained_models/contact_semantics.pt --pkl_file_path $POSA_dir/POSA_rp_poses/rp_aaron_posed_001_0_0.pkl --scene_name MPH16 --render 1 --viz 1 --use_cuda False

python src/gen_rand_samples.py --config cfg_files/contact.yaml --checkpoint_path $POSA_dir/trained_models/contact.pt --pkl_file_path $POSA_dir/POSA_rp_poses/rp_aaron_posed_001_0_0.pkl --render 1 --viz 1 --num_rand_samples 3 

Hello! This research is fantastic and I want to know about some datasets.
I know about PSI datasets. But this project is a little different.

1. data/sdf/MPH112_semantics.npy => Don't you know what this is for? I saw other sdf files.
2. mesh_ds => What is A,D,U,mesh_0? Is it scene mesh or body mesh? If it is a scene mesh, can you descreibe me about numbering? (ex. A_0.npz for MPH112)
3. data/train/x.npy => What is X? I guess the PROXE coordinate data.
4. data/train/x_semantics.npy => What is x_semantics? I guess the PROXE semantic data.
5. data/train/vertices.npy => Is it full body vertex? or parameter of smplx? or anything?
6. data/train/joints_can.npy => Is it body part like back, butt, thighs, foot and so on?
7. data/train/pkl_file_paths.json => Is it utilize for train code? I think it is not necessary file.
8. data/train/recording_name.json => Also, I think it is not necessary file.

I'm sorry for many questions. Thanks for publishing code and have a good day!
Best regards.