render_textured_meshes.ipynb 23.9 KB
Newer Older
facebook-github-bot's avatar
facebook-github-bot committed
1
2
3
4
{
 "cells": [
  {
   "cell_type": "code",
5
   "execution_count": null,
6
7
8
9
10
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "_Ip8kp4TfBLZ"
   },
facebook-github-bot's avatar
facebook-github-bot committed
11
12
13
14
15
16
17
   "outputs": [],
   "source": [
    "# Copyright (c) Facebook, Inc. and its affiliates. All rights reserved."
   ]
  },
  {
   "cell_type": "markdown",
18
19
20
21
   "metadata": {
    "colab_type": "text",
    "id": "kuXHJv44fBLe"
   },
facebook-github-bot's avatar
facebook-github-bot committed
22
23
24
25
26
27
28
29
30
31
32
33
34
   "source": [
    "# Render a textured mesh\n",
    "\n",
    "This tutorial shows how to:\n",
    "- load a mesh and textures from an `.obj` file. \n",
    "- set up a renderer \n",
    "- render the mesh \n",
    "- vary the rendering settings such as lighting and camera position\n",
    "- use the batching features of the pytorch3d API to render the mesh from different viewpoints"
   ]
  },
  {
   "cell_type": "markdown",
35
36
37
38
   "metadata": {
    "colab_type": "text",
    "id": "Bnj3THhzfBLf"
   },
facebook-github-bot's avatar
facebook-github-bot committed
39
   "source": [
40
41
42
43
44
45
46
47
48
49
    "## 0. Install and Import modules"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "okLalbR_g7NS"
   },
   "source": [
50
    "Ensure `torch` and `torchvision` are installed. If `pytorch3d` is not installed, install it using the following cell:"
facebook-github-bot's avatar
facebook-github-bot committed
51
52
53
54
   ]
  },
  {
   "cell_type": "code",
55
   "execution_count": null,
56
57
58
59
60
61
62
63
64
65
66
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 717
    },
    "colab_type": "code",
    "id": "musUWTglgxSB",
    "outputId": "16d1a1b2-3f7f-43ed-ca28-a4d236cc0572"
   },
   "outputs": [],
   "source": [
67
    "import os\n",
68
    "import sys\n",
69
    "import torch\n",
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
    "need_pytorch3d=False\n",
    "try:\n",
    "    import pytorch3d\n",
    "except ModuleNotFoundError:\n",
    "    need_pytorch3d=True\n",
    "if need_pytorch3d:\n",
    "    if torch.__version__.startswith(\"1.7\") and sys.platform.startswith(\"linux\"):\n",
    "        # We try to install PyTorch3D via a released wheel.\n",
    "        version_str=\"\".join([\n",
    "            f\"py3{sys.version_info.minor}_cu\",\n",
    "            torch.version.cuda.replace(\".\",\"\"),\n",
    "            f\"_pyt{torch.__version__[0:5:2]}\"\n",
    "        ])\n",
    "        !pip install pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html\n",
    "    else:\n",
    "        # We try to install PyTorch3D from source.\n",
Jeremy Reizenstein's avatar
Jeremy Reizenstein committed
86
    "        !curl -LO https://github.com/NVIDIA/cub/archive/1.10.0.tar.gz\n",
87
88
89
    "        !tar xzf 1.10.0.tar.gz\n",
    "        os.environ[\"CUB_HOME\"] = os.getcwd() + \"/cub-1.10.0\"\n",
    "        !pip install 'git+https://github.com/facebookresearch/pytorch3d.git@stable'"
90
91
92
93
   ]
  },
  {
   "cell_type": "code",
94
   "execution_count": null,
95
96
97
98
99
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "nX99zdoffBLg"
   },
facebook-github-bot's avatar
facebook-github-bot committed
100
101
102
103
104
105
106
107
   "outputs": [],
   "source": [
    "import os\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "from skimage.io import imread\n",
    "\n",
    "# Util function for loading meshes\n",
Nikhila Ravi's avatar
Nikhila Ravi committed
108
    "from pytorch3d.io import load_objs_as_meshes, load_obj\n",
facebook-github-bot's avatar
facebook-github-bot committed
109
110
    "\n",
    "# Data structures and functions for rendering\n",
111
    "from pytorch3d.structures import Meshes\n",
112
113
    "from pytorch3d.vis.plotly_vis import AxisArgs, plot_batch_individually, plot_scene\n",
    "from pytorch3d.vis.texture_vis import texturesuv_image_matplotlib\n",
facebook-github-bot's avatar
facebook-github-bot committed
114
115
    "from pytorch3d.renderer import (\n",
    "    look_at_view_transform,\n",
Georgia Gkioxari's avatar
Georgia Gkioxari committed
116
    "    FoVPerspectiveCameras, \n",
facebook-github-bot's avatar
facebook-github-bot committed
117
118
119
120
121
122
    "    PointLights, \n",
    "    DirectionalLights, \n",
    "    Materials, \n",
    "    RasterizationSettings, \n",
    "    MeshRenderer, \n",
    "    MeshRasterizer,  \n",
123
    "    SoftPhongShader,\n",
124
125
    "    TexturesUV,\n",
    "    TexturesVertex\n",
facebook-github-bot's avatar
facebook-github-bot committed
126
127
    ")\n",
    "\n",
128
    "# add path for demo utils functions \n",
facebook-github-bot's avatar
facebook-github-bot committed
129
130
    "import sys\n",
    "import os\n",
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
    "sys.path.append(os.path.abspath(''))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Lxmehq6Zhrzv"
   },
   "source": [
    "If using **Google Colab**, fetch the utils file for plotting image grids:"
   ]
  },
  {
   "cell_type": "code",
146
   "execution_count": null,
147
148
149
150
151
152
153
154
155
156
157
158
159
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 204
    },
    "colab_type": "code",
    "id": "HZozr3Pmho-5",
    "outputId": "be5eb60d-5f65-4db1-cca0-44ee68c8f5fd"
   },
   "outputs": [],
   "source": [
    "!wget https://raw.githubusercontent.com/facebookresearch/pytorch3d/master/docs/tutorials/utils/plot_image_grid.py\n",
    "from plot_image_grid import image_grid"
facebook-github-bot's avatar
facebook-github-bot committed
160
161
162
163
164
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
165
166
    "colab_type": "text",
    "id": "g4B62MzYiJUM"
facebook-github-bot's avatar
facebook-github-bot committed
167
168
   },
   "source": [
169
170
171
172
173
    "OR if running **locally** uncomment and run the following cell:"
   ]
  },
  {
   "cell_type": "code",
174
   "execution_count": null,
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "paJ4Im8ahl7O"
   },
   "outputs": [],
   "source": [
    "# from utils import image_grid"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "5jGq772XfBLk"
   },
   "source": [
    "### 1. Load a mesh and texture file\n",
facebook-github-bot's avatar
facebook-github-bot committed
193
194
195
    "\n",
    "Load an `.obj` file and it's associated `.mtl` file and create a **Textures** and **Meshes** object. \n",
    "\n",
196
    "**Meshes** is a unique datastructure provided in PyTorch3D for working with batches of meshes of different sizes. \n",
facebook-github-bot's avatar
facebook-github-bot committed
197
    "\n",
198
    "**TexturesUV** is an auxillary datastructure for storing vertex uv and texture maps for meshes. \n",
facebook-github-bot's avatar
facebook-github-bot committed
199
200
201
202
    "\n",
    "**Meshes** has several class methods which are used throughout the rendering pipeline."
   ]
  },
203
204
205
206
207
208
209
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "a8eU4zo5jd_H"
   },
   "source": [
Nikhila Ravi's avatar
Nikhila Ravi committed
210
211
    "If running this notebook using **Google Colab**, run the following cell to fetch the mesh obj and texture files and save it at the path `data/cow_mesh`:\n",
    "If running locally, the data is already available at the correct path. "
212
213
   ]
  },
facebook-github-bot's avatar
facebook-github-bot committed
214
215
  {
   "cell_type": "code",
216
   "execution_count": null,
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 578
    },
    "colab_type": "code",
    "id": "tTm0cVuOjb1W",
    "outputId": "6cd7e2ec-65e1-4dcc-99e8-c347bc504f0a"
   },
   "outputs": [],
   "source": [
    "!mkdir -p data/cow_mesh\n",
    "!wget -P data/cow_mesh https://dl.fbaipublicfiles.com/pytorch3d/data/cow_mesh/cow.obj\n",
    "!wget -P data/cow_mesh https://dl.fbaipublicfiles.com/pytorch3d/data/cow_mesh/cow.mtl\n",
    "!wget -P data/cow_mesh https://dl.fbaipublicfiles.com/pytorch3d/data/cow_mesh/cow_texture.png"
   ]
  },
  {
   "cell_type": "code",
236
   "execution_count": null,
237
238
239
240
241
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "gi5Kd0GafBLl"
   },
facebook-github-bot's avatar
facebook-github-bot committed
242
243
244
   "outputs": [],
   "source": [
    "# Setup\n",
Jeremy Reizenstein's avatar
Jeremy Reizenstein committed
245
246
247
248
249
    "if torch.cuda.is_available():\n",
    "    device = torch.device(\"cuda:0\")\n",
    "    torch.cuda.set_device(device)\n",
    "else:\n",
    "    device = torch.device(\"cpu\")\n",
facebook-github-bot's avatar
facebook-github-bot committed
250
251
252
253
254
255
    "\n",
    "# Set paths\n",
    "DATA_DIR = \"./data\"\n",
    "obj_filename = os.path.join(DATA_DIR, \"cow_mesh/cow.obj\")\n",
    "\n",
    "# Load obj file\n",
256
    "mesh = load_objs_as_meshes([obj_filename], device=device)"
facebook-github-bot's avatar
facebook-github-bot committed
257
258
259
260
   ]
  },
  {
   "cell_type": "markdown",
261
262
263
264
   "metadata": {
    "colab_type": "text",
    "id": "5APAQs6-fBLp"
   },
facebook-github-bot's avatar
facebook-github-bot committed
265
266
267
268
269
270
   "source": [
    "#### Let's visualize the texture map"
   ]
  },
  {
   "cell_type": "code",
271
   "execution_count": null,
272
273
274
275
276
277
278
279
280
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 428
    },
    "colab_type": "code",
    "id": "YipUhrIHfBLq",
    "outputId": "48987b1d-5cc1-4c2a-cb3c-713d64f6a38d"
   },
281
   "outputs": [],
facebook-github-bot's avatar
facebook-github-bot committed
282
283
   "source": [
    "plt.figure(figsize=(7,7))\n",
284
    "texture_image=mesh.textures.maps_padded()\n",
facebook-github-bot's avatar
facebook-github-bot committed
285
    "plt.imshow(texture_image.squeeze().cpu().numpy())\n",
286
    "plt.grid(\"off\");\n",
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
    "plt.axis(\"off\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "PyTorch3D has a built-in way to view the texture map with matplotlib along with the points on the map corresponding to vertices. There is also a method, texturesuv_image_PIL, to get a similar image which can be saved to a file."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.figure(figsize=(7,7))\n",
    "texturesuv_image_matplotlib(mesh.textures, subsample=None)\n",
    "plt.grid(\"off\");\n",
    "plt.axis(\"off\");"
facebook-github-bot's avatar
facebook-github-bot committed
307
308
309
310
   ]
  },
  {
   "cell_type": "markdown",
311
312
313
314
   "metadata": {
    "colab_type": "text",
    "id": "GcnG6XJ6fBLu"
   },
facebook-github-bot's avatar
facebook-github-bot committed
315
   "source": [
316
    "## 2. Create a renderer\n",
facebook-github-bot's avatar
facebook-github-bot committed
317
    "\n",
318
    "A renderer in PyTorch3D is composed of a **rasterizer** and a **shader** which each have a number of subcomponents such as a **camera** (orthographic/perspective). Here we initialize some of these components and use default values for the rest.\n",
facebook-github-bot's avatar
facebook-github-bot committed
319
320
321
322
323
324
    "\n",
    "In this example we will first create a **renderer** which uses a **perspective camera**, a **point light** and applies **phong shading**. Then we learn how to vary different components using the modular API.  "
   ]
  },
  {
   "cell_type": "code",
325
   "execution_count": null,
326
327
328
329
330
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "dX466mWnfBLv"
   },
facebook-github-bot's avatar
facebook-github-bot committed
331
332
   "outputs": [],
   "source": [
Georgia Gkioxari's avatar
Georgia Gkioxari committed
333
    "# Initialize a camera.\n",
334
335
336
    "# With world coordinates +Y up, +X left and +Z in, the front of the cow is facing the -Z direction. \n",
    "# So we move the camera by 180 in the azimuth direction so it is facing the front of the cow. \n",
    "R, T = look_at_view_transform(2.7, 0, 180) \n",
Georgia Gkioxari's avatar
Georgia Gkioxari committed
337
    "cameras = FoVPerspectiveCameras(device=device, R=R, T=T)\n",
facebook-github-bot's avatar
facebook-github-bot committed
338
339
340
    "\n",
    "# Define the settings for rasterization and shading. Here we set the output image to be of size\n",
    "# 512x512. As we are rendering images for visualization purposes only we will set faces_per_pixel=1\n",
Nikhila Ravi's avatar
Nikhila Ravi committed
341
342
343
344
    "# and blur_radius=0.0. We also set bin_size and max_faces_per_bin to None which ensure that \n",
    "# the faster coarse-to-fine rasterization method is used. Refer to rasterize_meshes.py for \n",
    "# explanations of these parameters. Refer to docs/notes/renderer.md for an explanation of \n",
    "# the difference between naive and coarse-to-fine rasterization. \n",
facebook-github-bot's avatar
facebook-github-bot committed
345
346
347
348
349
350
    "raster_settings = RasterizationSettings(\n",
    "    image_size=512, \n",
    "    blur_radius=0.0, \n",
    "    faces_per_pixel=1, \n",
    ")\n",
    "\n",
351
352
353
    "# Place a point light in front of the object. As mentioned above, the front of the cow is facing the \n",
    "# -z direction. \n",
    "lights = PointLights(device=device, location=[[0.0, 0.0, -3.0]])\n",
facebook-github-bot's avatar
facebook-github-bot committed
354
355
356
357
358
359
360
361
362
    "\n",
    "# Create a phong renderer by composing a rasterizer and a shader. The textured phong shader will \n",
    "# interpolate the texture uv coordinates for each vertex, sample from a texture image and \n",
    "# apply the Phong lighting model\n",
    "renderer = MeshRenderer(\n",
    "    rasterizer=MeshRasterizer(\n",
    "        cameras=cameras, \n",
    "        raster_settings=raster_settings\n",
    "    ),\n",
Nikhila Ravi's avatar
Nikhila Ravi committed
363
    "    shader=SoftPhongShader(\n",
facebook-github-bot's avatar
facebook-github-bot committed
364
365
366
367
368
369
370
371
372
    "        device=device, \n",
    "        cameras=cameras,\n",
    "        lights=lights\n",
    "    )\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
373
374
375
376
   "metadata": {
    "colab_type": "text",
    "id": "KyOY5qXvfBLz"
   },
facebook-github-bot's avatar
facebook-github-bot committed
377
   "source": [
378
    "## 3. Render the mesh"
facebook-github-bot's avatar
facebook-github-bot committed
379
380
381
382
   ]
  },
  {
   "cell_type": "markdown",
383
384
385
386
   "metadata": {
    "colab_type": "text",
    "id": "8VkRA4qJfBL0"
   },
facebook-github-bot's avatar
facebook-github-bot committed
387
388
389
390
391
392
   "source": [
    "The light is in front of the object so it is bright and the image has specular highlights."
   ]
  },
  {
   "cell_type": "code",
393
   "execution_count": null,
394
395
396
397
398
399
400
401
402
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 592
    },
    "colab_type": "code",
    "id": "gBLZH8iUfBL1",
    "outputId": "cc3cd3f0-189e-4497-ce47-e64b4da542e8"
   },
403
   "outputs": [],
facebook-github-bot's avatar
facebook-github-bot committed
404
405
406
407
   "source": [
    "images = renderer(mesh)\n",
    "plt.figure(figsize=(10, 10))\n",
    "plt.imshow(images[0, ..., :3].cpu().numpy())\n",
408
409
    "plt.grid(\"off\");\n",
    "plt.axis(\"off\");"
facebook-github-bot's avatar
facebook-github-bot committed
410
411
412
413
   ]
  },
  {
   "cell_type": "markdown",
414
415
416
417
   "metadata": {
    "colab_type": "text",
    "id": "k161XF3sfBL5"
   },
facebook-github-bot's avatar
facebook-github-bot committed
418
   "source": [
419
    "## 4. Move the light behind the object and re-render\n",
facebook-github-bot's avatar
facebook-github-bot committed
420
421
422
423
424
425
426
427
428
429
    "\n",
    "We can pass arbirary keyword arguments to the `rasterizer`/`shader` via the call to the `renderer` so the renderer does not need to be reinitialized if any of the settings change/\n",
    "\n",
    "In this case, we can simply update the location of the lights and pass them into the call to the renderer. \n",
    "\n",
    "The image is now dark as there is only ambient lighting, and there are no specular highlights."
   ]
  },
  {
   "cell_type": "code",
430
   "execution_count": null,
431
432
433
434
435
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "BdWkkeibfBL6"
   },
facebook-github-bot's avatar
facebook-github-bot committed
436
437
   "outputs": [],
   "source": [
438
    "# Now move the light so it is on the +Z axis which will be behind the cow. \n",
facebook-github-bot's avatar
facebook-github-bot committed
439
440
441
442
443
444
    "lights.location = torch.tensor([0.0, 0.0, +1.0], device=device)[None]\n",
    "images = renderer(mesh, lights=lights)"
   ]
  },
  {
   "cell_type": "code",
445
   "execution_count": null,
446
447
448
449
450
451
452
453
454
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 592
    },
    "colab_type": "code",
    "id": "UmV3j1YffBL9",
    "outputId": "2e8edca0-5bd8-4a2f-a160-83c4b0520123"
   },
455
   "outputs": [],
facebook-github-bot's avatar
facebook-github-bot committed
456
457
458
   "source": [
    "plt.figure(figsize=(10, 10))\n",
    "plt.imshow(images[0, ..., :3].cpu().numpy())\n",
459
460
    "plt.grid(\"off\");\n",
    "plt.axis(\"off\");"
facebook-github-bot's avatar
facebook-github-bot committed
461
462
463
464
   ]
  },
  {
   "cell_type": "markdown",
465
466
467
468
   "metadata": {
    "colab_type": "text",
    "id": "t93aVotMfBMB"
   },
facebook-github-bot's avatar
facebook-github-bot committed
469
   "source": [
470
    "## 5. Rotate the object, modify the material properties or light properties\n",
facebook-github-bot's avatar
facebook-github-bot committed
471
472
473
474
475
476
477
478
479
480
    "\n",
    "We can also change many other settings in the rendering pipeline. Here we:\n",
    "\n",
    "- change the **viewing angle** of the camera\n",
    "- change the **position** of the point light\n",
    "- change the **material reflectance** properties of the mesh"
   ]
  },
  {
   "cell_type": "code",
481
   "execution_count": null,
482
483
484
485
486
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4mYXYziefBMB"
   },
facebook-github-bot's avatar
facebook-github-bot committed
487
488
   "outputs": [],
   "source": [
489
490
    "# Rotate the object by increasing the elevation and azimuth angles\n",
    "R, T = look_at_view_transform(dist=2.7, elev=10, azim=-150)\n",
Georgia Gkioxari's avatar
Georgia Gkioxari committed
491
    "cameras = FoVPerspectiveCameras(device=device, R=R, T=T)\n",
facebook-github-bot's avatar
facebook-github-bot committed
492
    "\n",
493
494
    "# Move the light location so the light is shining on the cow's face.  \n",
    "lights.location = torch.tensor([[2.0, 2.0, -2.0]], device=device)\n",
facebook-github-bot's avatar
facebook-github-bot committed
495
496
497
498
499
500
501
502
503
504
505
506
507
508
    "\n",
    "# Change specular color to green and change material shininess \n",
    "materials = Materials(\n",
    "    device=device,\n",
    "    specular_color=[[0.0, 1.0, 0.0]],\n",
    "    shininess=10.0\n",
    ")\n",
    "\n",
    "# Re render the mesh, passing in keyword arguments for the modified components.\n",
    "images = renderer(mesh, lights=lights, materials=materials, cameras=cameras)"
   ]
  },
  {
   "cell_type": "code",
509
   "execution_count": null,
510
511
512
513
514
515
516
517
518
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 592
    },
    "colab_type": "code",
    "id": "rHIxIfh5fBME",
    "outputId": "1ca2d337-2983-478f-b3c9-d64b84ba1a31"
   },
519
   "outputs": [],
facebook-github-bot's avatar
facebook-github-bot committed
520
521
522
   "source": [
    "plt.figure(figsize=(10, 10))\n",
    "plt.imshow(images[0, ..., :3].cpu().numpy())\n",
523
524
    "plt.grid(\"off\");\n",
    "plt.axis(\"off\");"
facebook-github-bot's avatar
facebook-github-bot committed
525
526
527
528
   ]
  },
  {
   "cell_type": "markdown",
529
530
531
532
   "metadata": {
    "colab_type": "text",
    "id": "17c4xmtyfBMH"
   },
facebook-github-bot's avatar
facebook-github-bot committed
533
   "source": [
534
    "## 6. Batched Rendering\n",
facebook-github-bot's avatar
facebook-github-bot committed
535
    "\n",
536
    "One of the core design choices of the PyTorch3D API is to support **batched inputs for all components**. \n",
facebook-github-bot's avatar
facebook-github-bot committed
537
538
539
540
541
    "The renderer and associated components can take batched inputs and **render a batch of output images in one forward pass**. We will now use this feature to render the mesh from many different viewpoints.\n"
   ]
  },
  {
   "cell_type": "code",
542
   "execution_count": null,
543
544
545
546
547
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "CDQKebNNfBMI"
   },
facebook-github-bot's avatar
facebook-github-bot committed
548
549
550
551
552
553
554
555
556
557
558
   "outputs": [],
   "source": [
    "# Set batch size - this is the number of different viewpoints from which we want to render the mesh.\n",
    "batch_size = 20\n",
    "\n",
    "# Create a batch of meshes by repeating the cow mesh and associated textures. \n",
    "# Meshes has a useful `extend` method which allows us do this very easily. \n",
    "# This also extends the textures. \n",
    "meshes = mesh.extend(batch_size)\n",
    "\n",
    "# Get a batch of viewing angles. \n",
559
560
    "elev = torch.linspace(0, 180, batch_size)\n",
    "azim = torch.linspace(-180, 180, batch_size)\n",
facebook-github-bot's avatar
facebook-github-bot committed
561
562
563
564
565
    "\n",
    "# All the cameras helper methods support mixed type inputs and broadcasting. So we can \n",
    "# view the camera from the same distance and specify dist=2.7 as a float,\n",
    "# and then specify elevation and azimuth angles for each viewpoint as tensors. \n",
    "R, T = look_at_view_transform(dist=2.7, elev=elev, azim=azim)\n",
Georgia Gkioxari's avatar
Georgia Gkioxari committed
566
    "cameras = FoVPerspectiveCameras(device=device, R=R, T=T)\n",
facebook-github-bot's avatar
facebook-github-bot committed
567
    "\n",
568
569
    "# Move the light back in front of the cow which is facing the -z direction.\n",
    "lights.location = torch.tensor([[0.0, 0.0, -3.0]], device=device)"
facebook-github-bot's avatar
facebook-github-bot committed
570
571
572
573
   ]
  },
  {
   "cell_type": "code",
574
   "execution_count": null,
575
576
577
578
579
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "gyYJCwEDfBML"
   },
facebook-github-bot's avatar
facebook-github-bot committed
580
581
582
583
   "outputs": [],
   "source": [
    "# We can pass arbirary keyword arguments to the rasterizer/shader via the renderer\n",
    "# so the renderer does not need to be reinitialized if any of the settings change.\n",
584
    "images = renderer(meshes, cameras=cameras, lights=lights)"
facebook-github-bot's avatar
facebook-github-bot committed
585
586
587
588
   ]
  },
  {
   "cell_type": "code",
589
   "execution_count": null,
590
   "metadata": {},
591
   "outputs": [],
facebook-github-bot's avatar
facebook-github-bot committed
592
593
594
   "source": [
    "image_grid(images.cpu().numpy(), rows=4, cols=5, rgb=True)"
   ]
595
  },
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 7. Plotly visualization \n",
    "If you only want to visualize a mesh, you don't really need to use a differentiable renderer - instead we support plotting of Meshes with plotly. For these Meshes, we use TexturesVertex to define a texture for the rendering.\n",
    "`plot_meshes` creates a Plotly figure with a trace for each Meshes object. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "verts, faces_idx, _ = load_obj(obj_filename)\n",
    "faces = faces_idx.verts_idx\n",
    "\n",
    "# Initialize each vertex to be white in color.\n",
    "verts_rgb = torch.ones_like(verts)[None]  # (1, V, 3)\n",
    "textures = TexturesVertex(verts_features=verts_rgb.to(device))\n",
    "\n",
    "# Create a Meshes object\n",
    "mesh = Meshes(\n",
    "    verts=[verts.to(device)],   \n",
    "    faces=[faces.to(device)],\n",
    "    textures=textures\n",
    ")\n",
    "\n",
    "# Render the plotly figure\n",
626
627
628
629
630
    "fig = plot_scene({\n",
    "    \"subplot1\": {\n",
    "        \"cow_mesh\": mesh\n",
    "    }\n",
    "})\n",
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# use Plotly's default colors (no texture)\n",
    "mesh = Meshes(\n",
    "    verts=[verts.to(device)],   \n",
    "    faces=[faces.to(device)]\n",
    ")\n",
    "\n",
    "# Render the plotly figure\n",
647
648
649
650
651
652
    "fig = plot_scene({\n",
    "    \"subplot1\": {\n",
    "        \"cow_mesh\": mesh\n",
    "    }\n",
    "})\n",
    "fig.show()"
653
654
655
656
657
658
659
660
661
662
663
664
665
666
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a batch of meshes, and offset one to prevent overlap\n",
    "mesh_batch = Meshes(\n",
    "    verts=[verts.to(device), (verts + 2).to(device)],   \n",
    "    faces=[faces.to(device), faces.to(device)]\n",
    ")\n",
    "\n",
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
    "# plot mesh batch in the same trace\n",
    "fig = plot_scene({\n",
    "    \"subplot1\": {\n",
    "        \"cow_mesh_batch\": mesh_batch\n",
    "    }\n",
    "})\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot batch of meshes in different traces\n",
    "fig = plot_scene({\n",
    "    \"subplot1\": {\n",
    "        \"cow_mesh1\": mesh_batch[0],\n",
    "        \"cow_mesh2\": mesh_batch[1]\n",
    "    }\n",
    "})\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plot batch of meshes in different subplots\n",
    "fig = plot_scene({\n",
    "    \"subplot1\": {\n",
    "        \"cow_mesh1\": mesh_batch[0]\n",
    "    },\n",
    "    \"subplot2\":{\n",
    "        \"cow_mesh2\": mesh_batch[1]\n",
    "    }\n",
    "})\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For batches, we can also use `plot_batch_individually` to avoid constructing the scene dictionary ourselves."
715
716
717
718
719
720
721
722
723
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# extend the batch to have 4 meshes\n",
724
    "mesh_4 = mesh_batch.extend(2)\n",
725
726
    "\n",
    "# visualize the batch in different subplots, 2 per row\n",
727
    "fig = plot_batch_individually(mesh_4)\n",
728
    "# we can update the figure height and width\n",
729
    "fig.update_layout(height=1000, width=500)\n",
730
731
732
733
734
735
736
    "fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
737
    "We can also modify the axis arguments and axis backgrounds in both functions. "
738
739
740
741
742
743
744
745
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
746
747
748
749
750
751
752
753
754
    "fig2 = plot_scene({\n",
    "    \"cow_plot1\": {\n",
    "        \"cows\": mesh_batch\n",
    "    }\n",
    "},\n",
    "    xaxis={\"backgroundcolor\":\"rgb(200, 200, 230)\"},\n",
    "    yaxis={\"backgroundcolor\":\"rgb(230, 200, 200)\"},\n",
    "    zaxis={\"backgroundcolor\":\"rgb(200, 230, 200)\"}, \n",
    "    axis_args=AxisArgs(showgrid=True))\n",
755
756
757
    "fig2.show()"
   ]
  },
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig3 = plot_batch_individually(\n",
    "    mesh_4, \n",
    "    ncols=2,\n",
    "    subplot_titles = [\"cow1\", \"cow2\", \"cow3\", \"cow4\"], # customize subplot titles\n",
    "    xaxis={\"backgroundcolor\":\"rgb(200, 200, 230)\"},\n",
    "    yaxis={\"backgroundcolor\":\"rgb(230, 200, 200)\"},\n",
    "    zaxis={\"backgroundcolor\":\"rgb(200, 230, 200)\"}, \n",
    "    axis_args=AxisArgs(showgrid=True))\n",
    "fig3.show()"
   ]
  },
775
776
777
778
779
780
781
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "t3qphI1ElUb5"
   },
   "source": [
782
783
    "## 8. Conclusion\n",
    "In this tutorial we learnt how to **load** a textured mesh from an obj file, initialize a PyTorch3D datastructure called **Meshes**, set up an **Renderer** consisting of a **Rasterizer** and a **Shader**, and modify several components of the rendering pipeline. We also learned how to render Meshes in Plotly figures."
784
   ]
facebook-github-bot's avatar
facebook-github-bot committed
785
786
787
  }
 ],
 "metadata": {
788
  "accelerator": "GPU",
Nikhila Ravi's avatar
Nikhila Ravi committed
789
  "anp_metadata": {
790
   "path": "notebooks/render_textured_meshes.ipynb"
Nikhila Ravi's avatar
Nikhila Ravi committed
791
  },
facebook-github-bot's avatar
facebook-github-bot committed
792
793
794
795
796
797
798
799
  "bento_stylesheets": {
   "bento/extensions/flow/main.css": true,
   "bento/extensions/kernel_selector/main.css": true,
   "bento/extensions/kernel_ui/main.css": true,
   "bento/extensions/new_kernel/main.css": true,
   "bento/extensions/system_usage/main.css": true,
   "bento/extensions/theme/main.css": true
  },
800
801
802
803
  "colab": {
   "name": "render_textured_meshes.ipynb",
   "provenance": []
  },
Nikhila Ravi's avatar
Nikhila Ravi committed
804
805
806
  "disseminate_notebook_info": {
   "backup_notebook_id": "569222367081034"
  },
facebook-github-bot's avatar
facebook-github-bot committed
807
  "kernelspec": {
808
   "display_name": "pytorch3d_etc (local)",
facebook-github-bot's avatar
facebook-github-bot committed
809
   "language": "python",
810
   "name": "pytorch3d_etc_local"
facebook-github-bot's avatar
facebook-github-bot committed
811
812
813
814
815
816
817
818
819
820
821
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
822
   "version": "3.7.5+"
facebook-github-bot's avatar
facebook-github-bot committed
823
824
825
  }
 },
 "nbformat": 4,
826
 "nbformat_minor": 1
facebook-github-bot's avatar
facebook-github-bot committed
827
}