-
Notifications
You must be signed in to change notification settings - Fork 2
/
vis_pointdae.py
487 lines (436 loc) · 18.8 KB
/
vis_pointdae.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
import os
import numpy as np
from pytorch3d.io import load_obj, save_obj
from pytorch3d.io import IO
from pytorch3d.io.ply_io import _save_ply
from pytorch3d.structures import Meshes
from pytorch3d.ops import sample_points_from_meshes
import torch
import math
import ipdb
# from pyntcloud import PyntCloud
# import pandas as pd
from PIL import Image, ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True
from pyntcloud import PyntCloud
import pandas as pd
import os
import torch
import matplotlib.pyplot as plt
# Util function for loading meshes
from pytorch3d.io import load_objs_as_meshes, load_obj
from pytorch3d.structures import Pointclouds
from pytorch3d.vis.plotly_vis import AxisArgs, plot_batch_individually, plot_scene
# Data structures and functions for rendering
from pytorch3d.structures import Meshes
from pytorch3d.vis.plotly_vis import AxisArgs, plot_batch_individually, plot_scene
from pytorch3d.vis.texture_vis import texturesuv_image_matplotlib
from pytorch3d.renderer import (
look_at_view_transform,
FoVPerspectiveCameras,
PointLights,
DirectionalLights,
Materials,
RasterizationSettings,
MeshRenderer,
MeshRasterizer,
SoftPhongShader,
TexturesUV,
TexturesVertex
)
from pytorch3d.renderer import (
look_at_view_transform,
FoVOrthographicCameras,
PointsRasterizationSettings,
PointsRenderer,
PulsarPointsRenderer,
PointsRasterizer,
AlphaCompositor,
NormWeightedCompositor
)
from matplotlib import pyplot as plt, colors
# add path for demo utils functions
import sys
import os
sys.path.append(os.path.abspath(''))
shapenet_v2_path = '/home/ssddata/shapenet/shapenet_v2/ShapeNetCore.v2/'
save_path = '/home/yabin/syn_project/point_cloud/Point-MAE/data/ShapeNet55-34/shapenet_pc_ours_with_normal/'
# ## 在数据处理的时候不！需要 （）
# def pc_normalize(pc):
# centroid = np.mean(pc, axis=0)
# pc = pc - centroid
# m = np.max(np.sqrt(np.sum(pc**2, axis=1)))
# pc = pc / m
# return pc
from pytorch3d.renderer import TexturesVertex
def tri_with_center_and_normal(center, normal, r):
normal = normal.float()
random_vector = torch.tensor([0.5, 0.5, 0.5]).float() # 1 * 3
normal = normal / torch.norm(normal, dim=-1, keepdim=True)
if (normal == random_vector).int().sum().item() == 0:
random_vector = torch.tensor([1, 0, 0]).float() # 1 * 3
perpendicular_vector = torch.cross(normal, random_vector, dim=-1)
perpendicular_vector = perpendicular_vector / torch.norm(perpendicular_vector, dim=-1, keepdim=True)
rotation1 = rotation_vector_around_vector(perpendicular_vector, normal, torch.tensor([math.pi * 2.0 / 3.0]))
rotation2 = rotation_vector_around_vector(perpendicular_vector, normal, torch.tensor([math.pi * 4.0 / 3.0]))
return center + perpendicular_vector * r, center + rotation1 * r, center + rotation2 * r
def rotation_vector_around_vector(input_vector, anchor_vector, rotation_degree):
# input_vector: torch size 3
c = torch.cos(rotation_degree)
s = torch.sin(rotation_degree)
C = 1.0 - c
Q = torch.zeros(3, 3)
Q[0, 0] = anchor_vector[0] * anchor_vector[0] * C + c
Q[0, 1] = anchor_vector[1] * anchor_vector[0] * C - anchor_vector[2] * s
Q[0, 2] = anchor_vector[2] * anchor_vector[0] * C + anchor_vector[1] * s
Q[1, 0] = anchor_vector[1] * anchor_vector[0] * C + anchor_vector[2] * s
Q[1, 1] = anchor_vector[1] * anchor_vector[1] * C + c
Q[1, 2] = anchor_vector[2] * anchor_vector[1] * C - anchor_vector[0] * s
Q[2, 0] = anchor_vector[0] * anchor_vector[2] * C - anchor_vector[1] * s
Q[2, 1] = anchor_vector[2] * anchor_vector[1] * C + anchor_vector[0] * s
Q[2, 2] = anchor_vector[2] * anchor_vector[2] * C + c
# expand_input = input_vector.view(3, 1).expand(3,3)
# return (expand_input * Q).sum(1)
# output = torch.zeros(3)
# output[0] = input_vector[0] * Q[0, 0] + input_vector[0] * Q[0,1] + input_vector[0] * Q[0,2]
# output[1] = input_vector[1] * Q[1, 0] + input_vector[1] * Q[1, 1] + input_vector[1] * Q[1, 2]
# output[2] = input_vector[2] * Q[2, 0] + input_vector[2] * Q[2, 1] + input_vector[2] * Q[2, 2]
output = torch.matmul(Q.float(), input_vector.view(3,1).float())
return output.view(3)
def generate_surfel_mesh(points, normals):
# 1024 * 3.
# for each points, calculate its three tri neighbor.
verts = torch.rand(points.size(0) * 3, points.size(1))
for i in range(points.size(0)):
tri_one, tri_two, tri_thr = tri_with_center_and_normal(points[i], normals[i], r=0.005)
verts[i * 3] = tri_one
verts[i * 3 + 1] = tri_two
verts[i * 3 + 2] = tri_thr
faces = torch.rand(points.size())
for i in range(points.size(0)):
faces[i, 0] = i * 3
faces[i, 1] = i * 3 + 1
faces[i, 2] = i * 3 + 2
faces = faces.long()
return verts, faces
# perpendicular_vector = torch.tensor([0,1,0])
# normal = torch.tensor([0,0,1])
# rotation1 = rotation_vector_around_vector(perpendicular_vector, normal, torch.tensor([math.pi * 2.0 / 3.0]))
# rotation2 = rotation_vector_around_vector(perpendicular_vector, normal, torch.tensor([math.pi * 4.0 / 3.0]))
# print(rotation1)
# print(rotation2)
# Initialize a camera.
# With world coordinates +Y up, +X left and +Z in, the front of the cow is facing the -Z direction.
# So we move the camera by 180 in the azimuth direction so it is facing the front of the cow.
R, T = look_at_view_transform(2.7, 0, 180)
cameras = FoVPerspectiveCameras( R=R, T=T)
# Define the settings for rasterization and shading. Here we set the output image to be of size
# 512x512. As we are rendering images for visualization purposes only we will set faces_per_pixel=1
# and blur_radius=0.0. We also set bin_size and max_faces_per_bin to None which ensure that
# the faster coarse-to-fine rasterization method is used. Refer to rasterize_meshes.py for
# explanations of these parameters. Refer to docs/notes/renderer.md for an explanation of
# the difference between naive and coarse-to-fine rasterization.
raster_settings = RasterizationSettings(
image_size=512,
blur_radius=0.0,
faces_per_pixel=1,
)
# Place a point light in front of the object. As mentioned above, the front of the cow is facing the
# -z direction.
lights = PointLights(location=[[0.0, 0.0, -3.0]])
# Create a Phong renderer by composing a rasterizer and a shader. The textured Phong shader will
# interpolate the texture uv coordinates for each vertex, sample from a texture image and
# apply the Phong lighting model
renderer = MeshRenderer(
rasterizer=MeshRasterizer(
cameras=cameras,
raster_settings=raster_settings
),
shader=SoftPhongShader(
cameras=cameras,
lights=lights
)
)
# shapenet_path = shapenet_v2_path
# def extract_point_normal(obj_path, with_normal=True, point_number=1024):
# try:
# verts, faces, _ = load_obj(obj_path)
# except:
# log = open(os.path.join('log.txt'), 'a')
# log.write('\n-------------------------------------------\n')
# log.write(obj_path)
# log.close()
# print(obj_path)
# return np.array([0])
#
# test_mesh = Meshes(verts=[verts], faces=[faces.verts_idx])
# sample_test = sample_points_from_meshes(test_mesh, point_number, return_normals=with_normal) # float tensor
#
# verts, faces = generate_surfel_mesh(sample_test[0][0], sample_test[1][0])
#
# # # assign constant color texture
# color = torch.zeros(1, verts.size(0), 3).float()
# color[:, :1000, 0] = 1
# color[:, 1000:2000, 1] = 0
# color[:, 2000:, 2] = 0
# print(color.size())
# print(color)
# test_mesh = Meshes(verts=[verts], faces=[faces]) # textures=TexturesVertex(verts_features=color))
# test_mesh.textures = TexturesVertex(verts_features=color)
#
# IO().save_mesh(test_mesh, 'test.ply', binary=False, colors_as_uint8=True)
# point_cloud = extract_point_normal(obj_file_name)
def generate_surfel_cloud(points, normals, root, name, normal_angle):
verts, faces = generate_surfel_mesh(points, normals)
# # assign constant color texture
color = torch.zeros(1, verts.size(0), 3).float()
## assign color accroding to normal angle.
normal_angle[normal_angle > 30] = 80
normal_angle[normal_angle < 30] = 0
# normal_angle = normal_angle.int()
cmap = plt.cm.cool
norm = colors.Normalize(vmin=0.0, vmax=80.0)
color = torch.from_numpy(cmap(norm(normal_angle)))[:,:3].repeat(1,3).reshape(1,-1, 3)
# color[:, :1000, 0] = 1
# color[:, 1000:2000, 1] = 0
# color[:, 2000:, 2] = 0
# print(color.size())
# print(color)
test_mesh = Meshes(verts=[verts], faces=[faces]) # textures=TexturesVertex(verts_features=color))
test_mesh.textures = TexturesVertex(verts_features=color)
IO().save_mesh(test_mesh, root + name + '_surfel.ply', binary=False, colors_as_uint8=True)
# Initialize a camera.
R_point, T_point = look_at_view_transform(20, 10, 0)
cameras_point = FoVOrthographicCameras(R=R_point, T=T_point, znear=0.01)
# Define the settings for rasterization and shading. Here we set the output image to be of size
# 512x512. As we are rendering images for visualization purposes only we will set faces_per_pixel=1
# and blur_radius=0.0. Refer to raster_points.py for explanations of these parameters.
raster_settings_point = PointsRasterizationSettings(
image_size=512,
radius = 0.003,
points_per_pixel = 10
)
# Create a points renderer by compositing points using an alpha compositor (nearer points
# are weighted more heavily). See [1] for an explanation.
rasterizer_point = PointsRasterizer(cameras=cameras_point, raster_settings=raster_settings_point)
renderer_point = PointsRenderer(
rasterizer=rasterizer_point,
compositor=AlphaCompositor()
)
def generate_point_cloud(points, root, name, point_dis):
color = torch.zeros(1, points.size(0), 3).float()
## assign color accroding to normal angle.
# normal_angle[normal_angle > 30] = 30
# normal_angle = normal_angle.int()
cmap = plt.cm.cool
norm = colors.Normalize(vmin=0.0, vmax=0.001)
color = torch.from_numpy(cmap(norm(point_dis)))[:,:3].reshape(-1, 3) * 255
color = color.long()
# print(points.size())
# print(color.size())
# print(color)
# df = pd.DataFrame(
# # same arguments that you are passing to visualize_pcl
# data=np.hstack((np.array(points), np.array(color))),
# columns=["x", "y", "z", "red", "green", "blue"])
# df[['red', 'green', 'blue']] = df[['red', 'green', 'blue']].astype(np.uint)
# point_cloud = Pointclouds(points=[points], features=[color])
# IO().save_point_cloud(point_cloud, root + name + '_reconstruct_pc.obj')
# image = renderer_point(point_cloud)
# cloud = PyntCloud(df)
# cloud.plot() ### we have colored points here.
fout = open(root + name + '_reconstruct_pc.obj', 'w')
for i in range(points.size(0)):
fout.write('v %f %f %f %d %d %d\n' % (
points[i, 0], points[i, 1], points[i, 2], color[i, 0], color[i, 1],
color[i, 2]))
fout.close()
# d = {'x': points[:, 0], 'y': points[:, 1], 'z': points[:, 2], 'red': color[:,0], 'green': color[:,1], 'blue':color[:, 2]}
# cloud = PyntCloud(pd.DataFrame(data=d))
# cloud.to_file(root + name + '_reconstruct_pc.ply')
# # assign constant color texture
# color[:, :1000, 0] = 1
# color[:, 1000:2000, 1] = 0
# color[:, 2000:, 2] = 0
# print(color.size())
# print(color)
# test_mesh = Meshes(verts=[verts], faces=[faces]) # textures=TexturesVertex(verts_features=color))
# test_mesh.textures = TexturesVertex(verts_features=color)
# IO().save_mesh(test_mesh, root + name + '_reconstruct_pc.ply', binary=False, colors_as_uint8=True)
def generate_point_cloud_full(points, root, name, point_dis):
color = torch.zeros(1, points.size(0), 3).float()
## assign color accroding to normal angle.
# normal_angle[normal_angle > 30] = 30
# normal_angle = normal_angle.int()
cmap = plt.cm.cool
norm = colors.Normalize(vmin=0.0, vmax=1.0)
color = torch.from_numpy(cmap(norm(point_dis)))[:,:3].reshape(-1, 3) * 255
color = color.long()
# print(points.size())
# print(color.size())
# print(color)
# df = pd.DataFrame(
# # same arguments that you are passing to visualize_pcl
# data=np.hstack((np.array(points), np.array(color))),
# columns=["x", "y", "z", "red", "green", "blue"])
# df[['red', 'green', 'blue']] = df[['red', 'green', 'blue']].astype(np.uint)
# point_cloud = Pointclouds(points=[points], features=[color])
# IO().save_point_cloud(point_cloud, root + name + '_reconstruct_pc.obj')
# image = renderer_point(point_cloud)
# cloud = PyntCloud(df)
# cloud.plot() ### we have colored points here.
d = {'x': points[:, 0], 'y': points[:, 1], 'z': points[:, 2], 'red': color[:,0], 'green': color[:,1], 'blue':color[:, 2]}
cloud = PyntCloud(pd.DataFrame(data=d))
cloud.to_file(root + name + '_gt_pc.ply')
def random_sample(pc, num):
pc = np.array(pc)
permutation = np.arange(pc.shape[0])
np.random.shuffle(permutation)
pc = pc[permutation[:num]]
return torch.from_numpy(pc)
def farthest_point_sample(xyz, npoint):
"""
Input:
xyz: pointcloud data, [B, N, C]
npoint: number of samples
Return:
centroids: sampled pointcloud index, [B, npoint]
"""
device = xyz.device
B, N, C = xyz.size()
centroids = torch.zeros(B, npoint, dtype=torch.long).to(device)
distance = torch.ones(B, N).to(device) * 1e10
centroid = torch.mean(xyz, dim=1, keepdim=True) # [B, 1, C]
dist = torch.sum((xyz - centroid) ** 2, -1)
farthest = torch.max(dist, -1)[1]
batch_indices = torch.arange(B, dtype=torch.long).to(device)
for i in range(npoint):
centroids[:, i] = farthest
centroid = xyz[batch_indices, farthest, :].view(B, 1, 3)
dist = torch.sum((xyz - centroid) ** 2, -1)
mask = dist < distance
distance[mask] = dist[mask]
farthest = torch.max(distance, -1)[1]
return centroids
def index_points(points, idx):
"""
Input:
points: input points data, [B, N, C]
idx: sample index data, [B, S]
Return:
new_points:, indexed points data, [B, S, C]
"""
# print(idx.shape)
device = points.device
B = points.shape[0]
view_shape = list(idx.shape)
view_shape[1:] = [1] * (len(view_shape) - 1)
repeat_shape = list(idx.shape)
repeat_shape[0] = 1
batch_indices = torch.arange(B, dtype=torch.long).to(device).view(view_shape).repeat(repeat_shape)
new_points = points[batch_indices, idx, :]
return new_points
expname_list = [
# "vis_pretrain_PointCAE_affine_r3_dropout_local",
# "vis_pretrain_PointCAE_affine_r3",
# "vis_pretrain_PointCAE_dropout_local",
# "vis_pretrain_PointCAE_rotate_z",
# "vis_pretrain_PointCAE_rotate",
# "vis_pretrain_PointCAE_reflection",
# "vis_pretrain_PointCAE_scale_nonorm",
# "vis_pretrain_PointCAE_shear",
# "vis_pretrain_PointCAE_translate",
# # affine transformation combination
# "vis_pretrain_PointCAE_affine_r3_dropout_patch",
# ####
# "vis_pretrain_PointCAE_clean",
# # add noise
# "vis_pretrain_PointCAE_add_global",
# "vis_pretrain_PointCAE_add_local",
# "vis_pretrain_PointCAE_jitter",
# ### drop points
# "vis_pretrain_PointCAE_dropout_global",
# "vis_pretrain_PointCAE_dropout_patch",
# "vis_pretrain_PointCAE_nonuniform_density",
###
# "vis_pretrain_PointCAE_dropout_local_occow",
# "vis_pretrain_PointCAE_affine_r3_occow"
# "vis_pretrain_PointCAE_transformer_dropout_patch_affine_r3_maskpatch_p0005_PointCAE_transformer_with_fc_center_p"
"vis_pretrain_PointCAE_affine_r3_dropout_local_4xlonger_Point_CAE_DGCNN_FCOnly"
# "vis_pretrain_PointCAE_transformer_dropout_patch_maskpatch_p0005_PointCAE_transformer_with_fc_center_p",
# "vis_pretrain_PointCAE_transformer_affine_r3_PointCAE_transformer_with_fc_center_p"
]
for expname in expname_list:
# root = './vis/' + expname + '/02691156_520/' %%% 500, 520, 225, 234, 296, 298, 66, 77
root = './vis/' + expname + '/03467517_77/'
obj_file_name = root + 'pointdae.pth.tar'
a = torch.load(obj_file_name)
input_pc = a['input_point'][0]
coarse = a['coarse'][0]
fine = a['fine']
idx = farthest_point_sample(fine, 1024)
fine = index_points(fine, idx)[0] #[B, S, C][0]
gt = a['gt'][0]
d = {'x': input_pc[:, 0], 'y': input_pc[:, 1], 'z': input_pc[:, 2]}
cloud = PyntCloud(pd.DataFrame(data=d))
cloud.to_file(root + 'inputpc.ply')
d = {'x': coarse[:, 0], 'y': coarse[:, 1], 'z': coarse[:, 2]}
cloud = PyntCloud(pd.DataFrame(data=d))
cloud.to_file(root + 'coarse.ply')
d = {'x': fine[:, 0], 'y': fine[:, 1], 'z': fine[:, 2]}
cloud = PyntCloud(pd.DataFrame(data=d))
cloud.to_file(root + 'fine.ply')
d = {'x': gt[:, 0], 'y': gt[:, 1], 'z': gt[:, 2]}
cloud = PyntCloud(pd.DataFrame(data=d))
cloud.to_file(root + 'gt.ply')
# output_surfels = a['output_surfels']
# full_rebuild_points = output_surfels[:, :3]
# full_rebuild_normal = output_surfels[:, 3:6]
# point_dis = output_surfels[:, 6]
# normal_angle = output_surfels[:, 7]
# full_input = output_surfels[:, 8:]
#
# # print(normal_angle.mean()) # mae, 28.91, masksurf:
# # mae: 1216; 974, 828, 671, 532
# # masksurl:1172; 930, 801, 647, 494
# print((normal_angle>10).int().sum())
# print((normal_angle>20).int().sum())
# print((normal_angle>30).int().sum())
# print((normal_angle>40).int().sum())
# print((normal_angle>50).int().sum())
#
# d = {'x': input_pc[:, 0], 'y': input_pc[:, 1], 'z': input_pc[:, 2]}
# cloud = PyntCloud(pd.DataFrame(data=d))
# cloud.to_file(root + name + '_inputpc.ply')
# #
# generate_point_cloud(full_rebuild_points, root, name, point_dis)
# generate_point_cloud_full(full_input, root, name, point_dis)
# generate_surfel_cloud(full_rebuild_points, full_rebuild_normal, root, name, normal_angle)
#
# name = 'masksurf' #
# obj_file_name = root + name + '.pth.tar'
# a = torch.load(obj_file_name)
# input_pc = a['input_vis_point']
# output_surfels = a['output_surfels']
# full_rebuild_points = output_surfels[:, :3]
# full_rebuild_normal = output_surfels[:, 3:6]
# point_dis = output_surfels[:, 6]
# normal_angle = output_surfels[:, 7]
# full_input = output_surfels[:, 8:]
#
# # print(normal_angle.mean()) # mae, 28.91, masksurf:
# # mae: 1216; 974, 828, 671, 532
# # masksurl:1172; 930, 801, 647, 494
# print((normal_angle>10).int().sum())
# print((normal_angle>20).int().sum())
# print((normal_angle>30).int().sum())
# print((normal_angle>40).int().sum())
# print((normal_angle>50).int().sum())
#
# d = {'x': input_pc[:, 0], 'y': input_pc[:, 1], 'z': input_pc[:, 2]}
# cloud = PyntCloud(pd.DataFrame(data=d))
# cloud.to_file(root + name + '_inputpc.ply')
# #
# generate_point_cloud(full_rebuild_points, root, name, point_dis)
# generate_point_cloud_full(full_input, root, name, point_dis)
# generate_surfel_cloud(full_rebuild_points, full_rebuild_normal, root, name, normal_angle)