forked from yt-project/yt
-
Notifications
You must be signed in to change notification settings - Fork 0
/
anari_volume_rendering.py
401 lines (321 loc) · 15.3 KB
/
anari_volume_rendering.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
import numpy as np
from yt.funcs import is_sequence, mylog
from yt.utilities.lib.image_samplers import (
ImageSampler,
)
from anari import *
anari_library_name = "ospray"
anari_device_subtype = "default"
anari_renderer_name = "default"
prefixes = {
lib.ANARI_SEVERITY_FATAL_ERROR : "FATAL",
lib.ANARI_SEVERITY_ERROR : "ERROR",
lib.ANARI_SEVERITY_WARNING : "WARNING",
lib.ANARI_SEVERITY_PERFORMANCE_WARNING : "PERFORMANCE",
lib.ANARI_SEVERITY_INFO : "INFO",
lib.ANARI_SEVERITY_DEBUG : "DEBUG"
}
renderer_enum_info = [("default", "default", "default")]
def anari_status(device, source, sourceType, severity, code, message):
print('[%s]: '%prefixes[severity]+message)
def get_renderer_enum_info(self, context):
global renderer_enum_info
return renderer_enum_info
status_handle = ffi.new_handle(anari_status) #something needs to keep this handle alive
class AnariSampler:
def __init__(self, camera, render_source, params):
self.params = params
self.init_image_sampler(camera, render_source)
self.aimage = self.image_sampler.aimage
# ANARI library
self.library = anariLoadLibrary(anari_library_name, status_handle)
if not self.library:
print("Error loading ANARI library")
return
# ANARI device
self.device = anariNewDevice(self.library, anari_device_subtype)
if not self.device:
print("ANARI device not initialized")
return
# ANARI renderer
self.renderer = anariNewRenderer(self.device, anari_renderer_name)
anariSetParameter(self.device, self.renderer, "ambientColor", ANARI_FLOAT32_VEC3, [1,1,1])
anariSetParameter(self.device, self.renderer, "ambientRadiance", ANARI_FLOAT32, 1)
anariSetParameter(self.device, self.renderer, "pixelSamples", ANARI_INT32, 8)
anariCommitParameters(self.device, self.renderer)
cam_pos = [ camera.position[0],
camera.position[1],
camera.position[2] ]
cam_dir = [ camera.focus[0]-camera.position[0],
camera.focus[1]-camera.position[1],
camera.focus[2]-camera.position[2] ]
cam_up = [ camera.north_vector[0],
camera.north_vector[1],
camera.north_vector[2] ]
# ANARI camera
if self.params["lens_type"] == 'plane-parallel':
self.camera = anariNewCamera(self.device, 'orthographic')
anariSetParameter(self.device, self.camera, 'position', ANARI_FLOAT32_VEC3, cam_pos)
anariSetParameter(self.device, self.camera, 'direction', ANARI_FLOAT32_VEC3, cam_dir)
anariSetParameter(self.device, self.camera, 'up', ANARI_FLOAT32_VEC3, cam_up)
anariSetParameter(self.device, self.camera, 'height', ANARI_FLOAT32, camera.width[1])
elif self.params["lens_type"] == 'perspective':
self.camera = anariNewCamera(self.device, 'perspective')
anariSetParameter(self.device, self.camera, 'position', ANARI_FLOAT32_VEC3, cam_pos)
anariSetParameter(self.device, self.camera, 'direction', ANARI_FLOAT32_VEC3, cam_dir)
anariSetParameter(self.device, self.camera, 'up', ANARI_FLOAT32_VEC3, cam_up)
# ...
if self.camera is None:
print("Camera initialization failed")
return
anariCommitParameters(self.device, self.camera)
# ANARI frame
self.frame = anariNewFrame(self.device)
anariSetParameter(self.device, self.frame, 'channel.color', ANARI_DATA_TYPE, ANARI_UFIXED8_VEC4)
anariSetParameter(self.device, self.frame, 'channel.depth', ANARI_DATA_TYPE, ANARI_FLOAT32)
width = self.aimage.shape[0]
height = self.aimage.shape[1]
anariSetParameter(self.device, self.frame, 'size', ANARI_UINT32_VEC2, [width, height])
anariSetParameter(self.device, self.frame, 'renderer', ANARI_OBJECT, self.renderer)
anariSetParameter(self.device, self.frame, 'camera', ANARI_OBJECT, self.camera)
anariCommitParameters(self.device, self.frame)
def render_frame(self, volume, tf, viewpoint):
if not self.frame:
print("render_frame: Frame invalid..")
return
# TODO: should probably do this on init?!
volume.create_world(self.device, tf)
anariSetParameter(self.device, self.frame, 'world', ANARI_OBJECT, volume.world)
anariCommitParameters(self.device, self.frame)
anariRenderFrame(self.device, self.frame)
anariFrameReady(self.device, self.frame, ANARI_WAIT)
width = self.aimage.shape[0]
height = self.aimage.shape[1]
void_pixels, frame_width, frame_height, frame_type = anariMapFrame(self.device, self.frame, 'channel.color')
if frame_width != width or frame_height != height:
print('something went wrong..')
return
unpacked_pixels = ffi.buffer(void_pixels, frame_width*frame_height*4)
pixels = np.array(unpacked_pixels).astype(np.float32)#*(1.0/255.0)
rect = pixels.reshape((width, height, 4))
anariUnmapFrame(self.device, self.frame, 'channel.color')
for x in range(width):
for y in range(height):
for c in range(4):
self.aimage[x, y, c] = rect[x, y, c]
def init_image_sampler(self, camera, render_source):
args = (
np.atleast_3d(self.params["vp_pos"]),
np.atleast_3d(self.params["vp_dir"]),
self.params["center"],
self.params["bounds"],
self.params["image"],
self.params["x_vec"],
self.params["y_vec"],
self.params["width"],
render_source.volume_method,
self.params["transfer_function"],
self.params["num_samples"],
)
kwargs = {
"lens_type": self.params["lens_type"],
}
if render_source.zbuffer is not None:
kwargs["zbuffer"] = render_source.zbuffer.z
args[4][:] = np.reshape(
render_source.zbuffer.rgba[:],
(camera.resolution[0], camera.resolution[1], 4),
)
else:
kwargs["zbuffer"] = np.ones(self.params["image"].shape[:2], "float64")
self.image_sampler = ImageSampler(*args, **kwargs)
class AnariVolume:
fields = None
log_fields = None
def __init__(self, ds, data_source):
self.ds = ds
self.data_source = data_source
self.min_level = 0
self.max_level = ds.index.max_level
if data_source is None:
data_source = ds.all_data()
def set_fields(self, fields, log_fields, no_ghost, force=False):
new_fields = self.data_source._determine_fields(fields)
regenerate_data = (
self.fields is None
or len(self.fields) != len(new_fields)
or self.fields != new_fields
or force
)
if not is_sequence(log_fields):
log_fields = [log_fields]
new_log_fields = list(log_fields)
self.fields = new_fields
if self.log_fields is not None and not regenerate_data:
flip_log = list(map(operator.ne, self.log_fields, new_log_fields))
else:
flip_log = [False] * len(new_log_fields)
self.log_fields = new_log_fields
def tf_to_rgb(self, tf, height, width):
rgb = np.zeros((height, width, 3), dtype="float32")
hvals = np.mgrid[tf.x_bounds[0] : tf.x_bounds[1] : height * 1j]
for i, f in enumerate(tf.funcs[:3]):
vals = np.interp(hvals, f.x, f.y)
rgb[:, :, i] = (vals[:, None] * 1.0).astype("float32")
rgb = rgb[::-1, :, :]
return rgb
def tf_to_alpha(self, tf, height, width):
alpha = np.zeros((height, width, 1), dtype="float32")
hvals = np.mgrid[tf.x_bounds[0] : tf.x_bounds[1] : height * 1j]
vals = np.interp(hvals, tf.alpha.x, tf.alpha.y)
alpha[:, :, 0] = (vals[:, None] * 1.0).astype("float32")
alpha = alpha[::-1, :, :]
return alpha
def set_array(self, device, obj, name, atype, count, arr):
(ptr, stride) = anariMapParameterArray1D(device, obj, name, atype, count)
ffi.memmove(ptr, ffi.from_buffer(arr), arr.size*arr.itemsize)
anariUnmapParameterArray(device, obj, name)
def create_world(self, device, tf):
self.world = anariNewWorld(device)
self.volume = anariNewVolume(device, 'transferFunction1D')
self.field = anariNewSpatialField(device, 'amr')
cellWidth = []
blockBounds = []
blockLevel = []
blockData = []
lvl_range = range(self.min_level, self.max_level + 1)
# cell bounds (in logical coords):
x_min = 10000000
y_min = 10000000
z_min = 10000000
x_max = 0
y_max = 0
z_max = 0
# eventually, according to the provisional specs, level-0
# will become the coarsest, but he devices out there who
# support amr (visionaray, ospray) still have this the
# other way around:
anari_level_0_is_finest = True
for level in lvl_range:
cw = 0.0
if anari_level_0_is_finest:
cw = 2**(self.max_level - level)
else:
cw = 2**level
cellWidth.append(cw)
grids = np.array(
[b for b, mask in self.data_source.blocks if b.Level == level]
)
if len(grids) == 0:
continue
# cell bounds (in logical coords) on this level:
level_x_min = 10000000
level_y_min = 10000000
level_z_min = 10000000
level_x_max = 0
level_y_max = 0
level_z_max = 0
for grid in grids:
lower = np.rint(grid.LeftEdge / grid.dds).astype("int32")
upper = np.rint(grid.RightEdge / grid.dds).astype("int32")
box_min = [ lower[0]*cw,
lower[1]*cw,
lower[2]*cw ]
box_max = [ upper[0]*cw,
upper[1]*cw,
upper[2]*cw ]
x_min = min(x_min, box_min[0])
y_min = min(y_min, box_min[1])
z_min = min(z_min, box_min[2])
x_max = max(x_max, box_max[0])
y_max = max(y_max, box_max[1])
z_max = max(z_max, box_max[2])
#print("level: ",level)
#print("box_min,box_max: ",box_min,box_max)
#print("lower,upper: ",lower,upper)
level_x_min = min(level_x_min, box_min[0])
level_y_min = min(level_y_min, box_min[1])
level_z_min = min(level_z_min, box_min[2])
level_x_max = max(level_x_max, box_max[0])
level_y_max = max(level_y_max, box_max[1])
level_z_max = max(level_z_max, box_max[2])
dims = upper - lower
blockBounds.append([lower, upper])
blockLevel.append(level)
data = np.asarray(grid[self.fields[0]], dtype=np.float32)
flat = np.zeros(dims[0] * dims[1] * dims[2], dtype=np.float32)
for x, l in enumerate(data):
for y, ll in enumerate(l):
for z, e in enumerate(ll):
index = x+y*dims[0]+z*dims[0]*dims[1]
if self.log_fields:
flat[index] = np.log10(e)
else:
flat[index] = e
blockData.append(np.asarray(flat, dtype=np.float32))
#print("level: ",level)
#print("bounds:")
#print(level_x_min, level_y_min, level_z_min)
#print(level_x_max, level_y_max, level_z_max)
#print(x_min, y_min, z_min)
#print(x_max, y_max, z_max)
vmin = tf.x_bounds[0]
vmax = tf.x_bounds[1]
levelcount = len(cellWidth)
blockcount = len(blockLevel)
npcellWidth = np.zeros([levelcount], dtype=np.float32)
npblockBounds = np.zeros([blockcount*6], dtype=np.int32)
npblockLevel = np.zeros([blockcount], dtype=np.int32)
dataArrays = []
for i in range(levelcount):
npcellWidth[i] = cellWidth[i]
for i in range(blockcount):
npblockBounds[i*6] = blockBounds[i][0][0]
npblockBounds[i*6+1] = blockBounds[i][0][1]
npblockBounds[i*6+2] = blockBounds[i][0][2]
npblockBounds[i*6+3] = blockBounds[i][1][0]-1
npblockBounds[i*6+4] = blockBounds[i][1][1]-1
npblockBounds[i*6+5] = blockBounds[i][1][2]-1
npblockLevel[i] = blockLevel[i]
width = blockBounds[i][1][0]-blockBounds[i][0][0]
height = blockBounds[i][1][1]-blockBounds[i][0][1]
depth = blockBounds[i][1][2]-blockBounds[i][0][2]
bd = anariNewArray3D(device, ffi.from_buffer(blockData[i]), ANARI_FLOAT32, width, height, depth)
dataArrays.append(bd)
npblockData = ffi.new('ANARIArray3D[]', dataArrays)
array = anariNewArray1D(device, npblockData, ANARI_ARRAY1D, len(npblockData))
# compute gridOrigin and gridSpacing:
domain_width = self.data_source.ds.domain_width
domain_center = self.data_source.ds.domain_center
gridOrigin = [ domain_center[0] - domain_width[0]*0.5,
domain_center[1] - domain_width[1]*0.5,
domain_center[2] - domain_width[2]*0.5 ]
gridSpacing = [ domain_width[0] / (x_max-x_min),
domain_width[1] / (y_max-y_min),
domain_width[2] / (z_max-z_min) ]
self.set_array(device, self.field, 'cellWidth', ANARI_FLOAT32, levelcount, npcellWidth)
self.set_array(device, self.field, 'block.bounds', ANARI_INT32_BOX3, blockcount, npblockBounds)
self.set_array(device, self.field, 'block.level', ANARI_INT32, blockcount, npblockLevel)
anariSetParameter(device, self.field, 'block.data', ANARI_ARRAY3D, array)
anariSetParameter(device, self.field, 'gridOrigin', ANARI_FLOAT32_VEC3, gridOrigin)
anariSetParameter(device, self.field, 'gridSpacing', ANARI_FLOAT32_VEC3, gridSpacing)
anariCommitParameters(device, self.field)
anariSetParameter(device, self.volume, 'value', ANARI_SPATIAL_FIELD, self.field)
anariSetParameter(device, self.volume, 'valueRange', ANARI_FLOAT32_BOX1, [vmin, vmax])
anariCommitParameters(device, self.volume)
tf_size = 256
rgb = self.tf_to_rgb(tf, tf_size, 1)
alpha = self.tf_to_alpha(tf, tf_size, 1)
color = np.zeros(tf_size*3,dtype=np.float32)
opacity = np.zeros(tf_size,dtype=np.float32)
for i in range(tf_size):
color[i*3] = rgb[i][0][0]
color[i*3+1] = rgb[i][0][1]
color[i*3+2] = rgb[i][0][2]
opacity[i] = alpha[i][0][0]
self.set_array(device, self.volume, 'color', ANARI_FLOAT32_VEC3, tf_size, color)
self.set_array(device, self.volume, 'opacity', ANARI_FLOAT32, tf_size, opacity)
volumes = ffi.new('ANARIVolume[]', [self.volume])
array = anariNewArray1D(device, volumes, ANARI_VOLUME, 1)
anariSetParameter(device, self.world, 'volume', ANARI_ARRAY1D, array)
anariCommitParameters(device, self.world)