-
Notifications
You must be signed in to change notification settings - Fork 140
/
Copy pathtest_g_buffer.py
215 lines (200 loc) · 8.17 KB
/
test_g_buffer.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
import pyredner
import redner
import numpy as np
import torch
import skimage.transform
# Optimize depth and normal of a teapot
# Use GPU if available
pyredner.set_use_gpu(torch.cuda.is_available())
# Set up the pyredner scene for rendering:
material_map, mesh_list, light_map = pyredner.load_obj('scenes/teapot.obj')
for _, mesh in mesh_list:
mesh.normals = pyredner.compute_vertex_normal(mesh.vertices, mesh.indices)
# Setup camera
cam = pyredner.Camera(position = torch.tensor([0.0, 30.0, 200.0]),
look_at = torch.tensor([0.0, 30.0, 0.0]),
up = torch.tensor([0.0, 1.0, 0.0]),
fov = torch.tensor([45.0]), # in degree
clip_near = 1e-2, # needs to > 0
resolution = (256, 256),
fisheye = False)
# Setup materials
material_id_map = {}
materials = []
count = 0
for key, value in material_map.items():
material_id_map[key] = count
count += 1
materials.append(value)
# Setup geometries
shapes = []
for mtl_name, mesh in mesh_list:
shapes.append(pyredner.Shape(\
vertices = mesh.vertices,
indices = mesh.indices,
uvs = mesh.uvs,
normals = mesh.normals,
material_id = material_id_map[mtl_name]))
# We don't setup any light source here
# Construct the scene
scene = pyredner.Scene(cam, shapes, materials, area_lights = [], envmap = None)
# Serialize the scene
# Here we specify the output channels as "depth", "shading_normal"
scene_args = pyredner.RenderFunction.serialize_scene(\
scene = scene,
num_samples = 16,
max_bounces = 0,
channels = [redner.channels.depth, redner.channels.shading_normal])
# Render the scene as our target image.
render = pyredner.RenderFunction.apply
# Render. The first argument is the seed for RNG in the renderer.
img = render(0, *scene_args)
# Save the images.
depth = img[:, :, 0]
normal = img[:, :, 1:4]
pyredner.imwrite(depth.cpu(),
'results/test_g_buffer/target_depth.exr')
pyredner.imwrite(depth.cpu(),
'results/test_g_buffer/target_depth.png', normalize = True)
pyredner.imwrite(normal.cpu(),
'results/test_g_buffer/target_normal.exr')
pyredner.imwrite(normal.cpu(),
'results/test_g_buffer/target_normal.png', normalize = True)
# Read the target image we just saved.
target_depth = pyredner.imread('results/test_g_buffer/target_depth.exr')
target_normal = pyredner.imread('results/test_g_buffer/target_normal.exr')
if pyredner.get_use_gpu():
target_depth = target_depth.cuda(device = pyredner.get_device())
target_normal = target_normal.cuda(device = pyredner.get_device())
# Perturb the teapot by a translation and a rotation to the object
translation_params = torch.tensor([0.1, -0.1, 0.1],
device = pyredner.get_device(), requires_grad=True)
translation = translation_params * 100.0
euler_angles = torch.tensor([0.1, -0.1, 0.1], requires_grad=True)
# These are the vertices we want to apply the transformation
shape0_vertices = shapes[0].vertices.clone()
shape1_vertices = shapes[1].vertices.clone()
# We can use pyredner.gen_rotate_matrix to generate 3x3 rotation matrices
rotation_matrix = pyredner.gen_rotate_matrix(euler_angles)
if pyredner.get_use_gpu():
rotation_matrix = rotation_matrix.cuda()
center = torch.mean(torch.cat([shape0_vertices, shape1_vertices]), 0)
# Shift the vertices to the center, apply rotation matrix,
# shift back to the original space
shapes[0].vertices = \
(shape0_vertices - center) @ torch.t(rotation_matrix) + \
center + translation
shapes[1].vertices = \
(shape1_vertices - center) @ torch.t(rotation_matrix) + \
center + translation
# Since we changed the vertices, we need to regenerate the shading normals
shapes[0].normals = pyredner.compute_vertex_normal(shapes[0].vertices, shapes[0].indices)
shapes[1].normals = pyredner.compute_vertex_normal(shapes[1].vertices, shapes[1].indices)
# We need to serialize the scene again to get the new arguments.
scene_args = pyredner.RenderFunction.serialize_scene(\
scene = scene,
num_samples = 16,
max_bounces = 0,
channels = [redner.channels.depth, redner.channels.shading_normal])
# Render the initial guess.
img = render(1, *scene_args)
depth = img[:, :, 0]
normal = img[:, :, 1:4]
# Save the images.
pyredner.imwrite(depth.cpu(),
'results/test_g_buffer/init_depth.png',
normalize = True)
pyredner.imwrite(depth.cpu(),
'results/test_g_buffer/init_normal.png',
normalize = True)
# Compute the difference and save the images.
diff_depth = torch.abs(target_depth - depth)
diff_normal = torch.abs(target_normal - normal)
pyredner.imwrite(diff_depth.cpu(),
'results/test_g_buffer/init_depth_diff.png')
pyredner.imwrite(diff_normal.cpu(),
'results/test_g_buffer/init_normal_diff.png')
# Optimize for triangle vertices.
optimizer = torch.optim.Adam([translation_params, euler_angles], lr=1e-2)
# Run 200 Adam iterations.
for t in range(200):
print('iteration:', t)
optimizer.zero_grad()
# Forward pass: apply the mesh operation and render the image.
translation = translation_params * 100.0
rotation_matrix = pyredner.gen_rotate_matrix(euler_angles)
if pyredner.get_use_gpu():
rotation_matrix = rotation_matrix.cuda(device = pyredner.get_device())
center = torch.mean(torch.cat([shape0_vertices, shape1_vertices]), 0)
shapes[0].vertices = \
(shape0_vertices - center) @ torch.t(rotation_matrix) + \
center + translation
shapes[1].vertices = \
(shape1_vertices - center) @ torch.t(rotation_matrix) + \
center + translation
shapes[0].normals = pyredner.compute_vertex_normal(shapes[0].vertices, shapes[0].indices)
shapes[1].normals = pyredner.compute_vertex_normal(shapes[1].vertices, shapes[1].indices)
scene_args = pyredner.RenderFunction.serialize_scene(\
scene = scene,
num_samples = 4, # We use less samples in the Adam loop.
max_bounces = 0,
channels = [redner.channels.depth, redner.channels.shading_normal])
# Important to use a different seed every iteration, otherwise the result
# would be biased.
img = render(t+1, *scene_args)
depth = img[:, :, 0]
normal = img[:, :, 1:4]
# Save the intermediate render.
pyredner.imwrite(depth.cpu(),
'results/test_g_buffer/iter_depth_{}.png'.format(t),
normalize = True)
pyredner.imwrite(normal.cpu(),
'results/test_g_buffer/iter_normal_{}.png'.format(t),
normalize = True)
# Compute the loss function. Here it is L2.
loss = (depth - target_depth).pow(2).sum() / 200.0 + (normal - target_normal).pow(2).sum()
print('loss:', loss.item())
# Backpropagate the gradients.
loss.backward()
# Print the gradients
print('translation_params.grad:', translation_params.grad)
print('euler_angles.grad:', euler_angles.grad)
# Take a gradient descent step.
optimizer.step()
# Print the current pose parameters.
print('translation:', translation)
print('euler_angles:', euler_angles)
# Render the final result.
scene_args = pyredner.RenderFunction.serialize_scene(\
scene = scene,
num_samples = 16,
max_bounces = 0,
channels = [redner.channels.depth, redner.channels.shading_normal])
img = render(202, *scene_args)
depth = img[:, :, 0]
normal = img[:, :, 1:4]
# Save the images.
pyredner.imwrite(depth.cpu(),
'results/test_g_buffer/final_depth.exr')
pyredner.imwrite(depth.cpu(),
'results/test_g_buffer/init_depth.png',
normalize = True)
pyredner.imwrite(normal.cpu(),
'results/test_g_buffer/final_normal.exr')
pyredner.imwrite(normal.cpu(),
'results/test_g_buffer/final_normal.png',
normalize = True)
diff_depth = torch.abs(target_depth - depth)
diff_normal = torch.abs(target_normal - normal)
pyredner.imwrite(diff_depth.cpu(),
'results/test_g_buffer/init_depth_diff.png')
pyredner.imwrite(diff_normal.cpu(),
'results/test_g_buffer/init_normal_diff.png')
# Convert the intermediate renderings to a video.
from subprocess import call
call(["ffmpeg", "-framerate", "24", "-i",
"results/test_g_buffer/iter_depth_%d.png", "-vb", "20M",
"results/test_g_buffer/out_depth.mp4"])
call(["ffmpeg", "-framerate", "24", "-i",
"results/test_g_buffer/iter_normal_%d.png", "-vb", "20M",
"results/test_g_buffer/out_normal.mp4"])