lix-shape-analysis / surfmnet Goto Github PK

Dean Jean, thanks for sharing your wonderful work. I've found the processed shot descriptors of each dataset mentioned in your paper, but I didn't find the corresponding mesh data. Would you please also share the original mesh data?

Hi Jean-Michel, thanks for sharing the code of your wonderful work. I've tried to run the training code but encountered the following error. Would you please help to figure out what's the problem? Thank your.

Traceback (most recent call last):
File "train_DFMnet.py", line 285, in
tf.app.run()
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 124, in run
_sys.exit(main(argv))
File "train_DFMnet.py", line 280, in main
run_training()
File "train_DFMnet.py", line 260, in run_training
feed_dict=feed_dict
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 895, in run
run_metadata_ptr)
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1128, in _run
feed_dict_tensor, options, run_metadata)
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1344, in _do_run
options, run_metadata)
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1363, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: flat indices[139, :] = [1, 1319] does not index into param (shape:[2,352,1500]).
[[Node: func_map_loss/GatherNd_2 = GatherNd[Tindices=DT_INT32, Tparams=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](func_map_loss/transpose_3, func_map_loss/concat_1)]]

Caused by op 'func_map_loss/GatherNd_2', defined at:
File "train_DFMnet.py", line 285, in
tf.app.run()
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/platform/app.py", line 124, in run
sys.exit(main(argv))
File "train_DFMnet.py", line 280, in main
run_training()
File "train_DFMnet.py", line 202, in run_training
target_evecs_trans, target_evals
File "/home/mmvc/mmvc-ny-local/mmvc-ny-local-002/Xiang_Li/SURFMNet/DFMnet.py", line 83, in dfmnet_model
F, G
File "/home/mmvc/mmvc-ny-local/mmvc-ny-local-002/Xiang_Li/SURFMNet/loss_DFMnet.py", line 190, in func_map_layer
source_evecs, source_evecs_trans))/2
File "/home/mmvc/mmvc-ny-local/mmvc-ny-local-002/Xiang_Li/SURFMNet/loss_DFMnet.py", line 101, in penalty_desc_commutativity
F = tf.gather_nd(F_trans, indices) # percent% of descriptors chosen
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/ops/gen_array_ops.py", line 2000, ingather_nd
"GatherNd", params=params, indices=indices, name=name)
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3160, in create_op
op_def=op_def)
File "/home/mmvc/anaconda2/envs/XL_py3_tf15/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1625, in init
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

InvalidArgumentError (see above for traceback): flat indices[139, :] = [1, 1319] does not index into param (shape: [2,352,1500]).
[[Node: func_map_loss/GatherNd_2 = GatherNd[Tindices=DT_INT32, Tparams=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](func_map_loss/transpose_3, func_map_loss/concat_1)]]

Thanks for your contribution for the shape analysis with the deep learning concepts,but when i check the scape_remesh dataset,i find the mesh051 is lost.can you upload it again? thanks for your reply.

Hello,
I am trying to run SURFMNet on my own dataset, and I have finished calculating shot descriptor and evecs. But I faced difficulties when calculating target_evecs_trans. I don't understand how to transpose the evecs with mass matrix correction.
Could you please explain it or share the pre-processing code? Thank you a lot!

Hi! I downloaded the scape_r dataset but it only has SHOT descriptors. Could you also provide the corresponding mesh data?

Hi, thanks for your great work. But when I train the Faust of remesh like the code you provided, I can't get the same accuracy as the GT you provided. The accuracy is about 20% lower, and the match I get exceeds the limit. How can I solve this problem?

Dear professor,
Thanks for your great work, could you share the code about texture transfer, looking forward to your reply.

Dear Author,

Currently, I am working with functional map. However, I encounter a problem about the data provided by the given link that I want to hear your answer. I notice that in your other works for example, GeomFmaps. You provide the MATLAB code to compute the eigenvalues, eigenvectors and transposed eigenvectors. I notice that when the transposed eigenvector is computed. It will multiply with the mass matrix. While, in the link you provide for the SURFMNet, I notice the transposed eigenvector doesn't multiply with the mass matrix. I want to know which process is correct. I also notice the .off files in the link you provide in this repo is different than other link, like GeomFmaps. Could you give me some hints.

Thanks a lot.

Best regards,
Dongliang

Dear Author,

Because I would like to train SURFMNet with other datasets, so it would be very helpful that you can provide the code to compute the SHOT descriptor. I have already tried using PCL to compute the SHOT descriptor. Since the SHOT descriptor has some hyperparameters such as search radius, I cannot reproduce the pre-computed SHOT descriptor you provided. If it is possible, maybe you can also provide me the hyperparameters for SHOT computation.

Thanks in advance.

Best regards,
Dongliang

Hi, thanks for your great work. when I run the train_DFMnet.py ，the error appears which will show in the follow and I cannot figure out.
InternalError (see above for traceback): tensorflow/core/kernels/cuda_solvers.cc:375: cuSolverDN call failed with status =7
[[Node: matrix_solve_ls_1/Cholesky = CholeskyT=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]
[[Node: Adam/update/_1642 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device_incarnation=1, tensor_name="edge_8234_Adam/update", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]
looking forward for your reply.