dragonlong / equi-pose Goto Github PK

thanks for your excellent work.
I get a problem when I test the model you provided.
Here is the error:
adj_idx: [array([[4, 1, 5]]), array([[0, 2]]), array([[1, 3]]), array([[2, 4, 8]]), array([[0, 3, 9]]), array([[ 0, 10]]), array([[11, 12]]), array([[12, 13]]), array([[ 3, 14]]), array([[ 4, 10, 14]]), array([[ 9, 5, 15]]), array([[ 6, 16]]), array([[ 6, 7, 17]]), array([[ 7, 18]]), array([[ 9, 8, 19]]), array([[10, 19, 16]]), array([[11, 15, 17]]), array([[12, 16, 18]]), array([[13, 17, 19]]), array([[14, 15, 18]])]
Traceback (most recent call last):
File "main.py", line 10, in
from datasets.dataset_parser import DatasetParser
File "/home/sh/jun/equi-pose/datasets/dataset_parser.py", line 18, in
from datasets.modelnet40_complete import ModelNet40Complete # complete
File "/home/sh/jun/equi-pose/datasets/modelnet40_complete.py", line 9, in
import vgtk.pc as pctk
File "/home/sh/jun/equi-pose/models/vgtk/init.py", line 5, in
from . import so3conv
File "/home/sh/jun/equi-pose/models/vgtk/so3conv/init.py", line 2, in
from .functional import *
File "/home/sh/jun/equi-pose/models/vgtk/so3conv/functional.py", line 387, in
Rs, R_idx, canonical_relative = fr.icosahedron_so3_trimesh(ANCHOR_PATH, GAMMA_SIZE)
File "/home/sh/jun/equi-pose/models/vgtk/functional/rotation.py", line 260, in icosahedron_so3_trimesh
R_adj = get_adjmatrix_trimesh(mesh, gsize)
File "/home/sh/jun/equi-pose/models/vgtk/functional/rotation.py", line 128, in get_adjmatrix_trimesh
face_adj = np.vstack(adj_idx).astype(np.int32)
File "<array_function internals>", line 6, in vstack
File "/home/sh/anaconda3/envs/jun_pvn3d/lib/python3.6/site-packages/numpy/core/shape_base.py", line 283, in vstack
return _nx.concatenate(arrs, 0)
File "<array_function internals>", line 6, in concatenate
ValueError: all the input array dimensions for the concatenation axis must match exactly, but along dimension 1, the array at index 0 has size 3 and the array at index 1 has size 2

The error may be due to 'vstack' receiving arrays of different dimensions.
Could you please give me some advice to solve this problem？
Thanks!

Hello! thanks for your sharing!
I want to evaluate the model on NOCS-REAL as Table 5 in your paper.
Can you let us know more detail about it in your README please?

Thank you!!

Hello author, thank you for releasing the code, but there seems to be one missing file? quaternion.py?
In addition, will the training and testing details on the NOCS dataset be available?
Since this work is the first self-supervised method for category-level OPE problem, the complete training and testing details will greatly contribute to the follow-up research!!!