Run in Google Colab# Copyright (c) Meta Platforms, Inc. and affiliates. All rights reserved.
In this tutorial we will learn the [x, y, z] position of a camera given a reference image using differentiable rendering.
We will first initialize a renderer with a starting position for the camera. We will then use this to generate an image, compute a loss with the reference image, and finally backpropagate through the entire pipeline to update the position of the camera.
This tutorial shows how to:
.obj fileCamera, Shader and Renderer,Ensure torch and torchvision are installed. If pytorch3d is not installed, install it using the following cell:
import os
import sys
import torch
need_pytorch3d=False
try:
import pytorch3d
except ModuleNotFoundError:
need_pytorch3d=True
if need_pytorch3d:
if torch.__version__.startswith("1.13.") and sys.platform.startswith("linux"):
# We try to install PyTorch3D via a released wheel.
pyt_version_str=torch.__version__.split("+")[0].replace(".", "")
version_str="".join([
f"py3{sys.version_info.minor}_cu",
torch.version.cuda.replace(".",""),
f"_pyt{pyt_version_str}"
])
!pip install fvcore iopath
!pip install --no-index --no-cache-dir pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/{version_str}/download.html
else:
# We try to install PyTorch3D from source.
!curl -LO https://github.com/NVIDIA/cub/archive/1.10.0.tar.gz
!tar xzf 1.10.0.tar.gz
os.environ["CUB_HOME"] = os.getcwd() + "/cub-1.10.0"
!pip install 'git+https://github.com/facebookresearch/[email protected]'
import os
import torch
import numpy as np
from tqdm.notebook import tqdm
import imageio
import torch.nn as nn
import torch.nn.functional as F
import matplotlib.pyplot as plt
from skimage import img_as_ubyte
# io utils
from pytorch3d.io import load_obj
# datastructures
from pytorch3d.structures import Meshes
# 3D transformations functions
from pytorch3d.transforms import Rotate, Translate
# rendering components
from pytorch3d.renderer import (
FoVPerspectiveCameras, look_at_view_transform, look_at_rotation,
RasterizationSettings, MeshRenderer, MeshRasterizer, BlendParams,
SoftSilhouetteShader, HardPhongShader, PointLights, TexturesVertex,
)
We will load an obj file and create a Meshes object. Meshes is a unique datastructure provided in PyTorch3D for working with batches of meshes of different sizes. It has several useful class methods which are used in the rendering pipeline.
If you are running this notebook locally after cloning the PyTorch3D repository, the mesh will already be available. If using Google Colab, fetch the mesh and save it at the path data/:
!mkdir -p data
!wget -P data https://dl.fbaipublicfiles.com/pytorch3d/data/teapot/teapot.obj
# Set the cuda device
if torch.cuda.is_available():
device = torch.device("cuda:0")
torch.cuda.set_device(device)
else:
device = torch.device("cpu")
# Load the obj and ignore the textures and materials.
verts, faces_idx, _ = load_obj("./data/teapot.obj")
faces = faces_idx.verts_idx
# Initialize each vertex to be white in color.
verts_rgb = torch.ones_like(verts)[None] # (1, V, 3)
textures = TexturesVertex(verts_features=verts_rgb.to(device))
# Create a Meshes object for the teapot. Here we have only one mesh in the batch.
teapot_mesh = Meshes(
verts=[verts.to(device)],
faces=[faces.to(device)],
textures=textures
)
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.
For optimizing the camera position we will use a renderer which produces a silhouette of the object only and does not apply any lighting or shading. We will also initialize another renderer which applies full Phong shading and use this for visualizing the outputs.
# Initialize a perspective camera.
cameras = FoVPerspectiveCameras(device=device)
# To blend the 100 faces we set a few parameters which control the opacity and the sharpness of
# edges. Refer to blending.py for more details.
blend_params = BlendParams(sigma=1e-4, gamma=1e-4)
# Define the settings for rasterization and shading. Here we set the output image to be of size
# 256x256. To form the blended image we use 100 faces for each pixel. 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=256,
blur_radius=np.log(1. / 1e-4 - 1.) * blend_params.sigma,
faces_per_pixel=100,
)
# Create a silhouette mesh renderer by composing a rasterizer and a shader.
silhouette_renderer = MeshRenderer(
rasterizer=MeshRasterizer(
cameras=cameras,
raster_settings=raster_settings
),
shader=SoftSilhouetteShader(blend_params=blend_params)
)
# We will also create a Phong renderer. This is simpler and only needs to render one face per pixel.
raster_settings = RasterizationSettings(
image_size=256,
blur_radius=0.0,
faces_per_pixel=1,
)
# We can add a point light in front of the object.
lights = PointLights(device=device, location=((2.0, 2.0, -2.0),))
phong_renderer = MeshRenderer(
rasterizer=MeshRasterizer(
cameras=cameras,
raster_settings=raster_settings
),
shader=HardPhongShader(device=device, cameras=cameras, lights=lights)
)
We will first position the teapot and generate an image. We use helper functions to rotate the teapot to a desired viewpoint. Then we can use the renderers to produce an image. Here we will use both renderers and visualize the silhouette and full shaded image.
The world coordinate system is defined as +Y up, +X left and +Z in. The teapot in world coordinates has the spout pointing to the left.
We defined a camera which is positioned on the positive z axis hence sees the spout to the right.
# Select the viewpoint using spherical angles
distance = 3 # distance from camera to the object
elevation = 50.0 # angle of elevation in degrees
azimuth = 0.0 # No rotation so the camera is positioned on the +Z axis.
# Get the position of the camera based on the spherical angles
R, T = look_at_view_transform(distance, elevation, azimuth, device=device)
# Render the teapot providing the values of R and T.
silhouette = silhouette_renderer(meshes_world=teapot_mesh, R=R, T=T)
image_ref = phong_renderer(meshes_world=teapot_mesh, R=R, T=T)
silhouette = silhouette.cpu().numpy()
image_ref = image_ref.cpu().numpy()
plt.figure(figsize=(10, 10))
plt.subplot(1, 2, 1)
plt.imshow(silhouette.squeeze()[..., 3]) # only plot the alpha channel of the RGBA image
plt.grid(False)
plt.subplot(1, 2, 2)
plt.imshow(image_ref.squeeze())
plt.grid(False)
Here we create a simple model class and initialize a parameter for the camera position.
class Model(nn.Module):
def __init__(self, meshes, renderer, image_ref):
super().__init__()
self.meshes = meshes
self.device = meshes.device
self.renderer = renderer
# Get the silhouette of the reference RGB image by finding all non-white pixel values.
image_ref = torch.from_numpy((image_ref[..., :3].max(-1) != 1).astype(np.float32))
self.register_buffer('image_ref', image_ref)
# Create an optimizable parameter for the x, y, z position of the camera.
self.camera_position = nn.Parameter(
torch.from_numpy(np.array([3.0, 6.9, +2.5], dtype=np.float32)).to(meshes.device))
def forward(self):
# Render the image using the updated camera position. Based on the new position of the
# camera we calculate the rotation and translation matrices
R = look_at_rotation(self.camera_position[None, :], device=self.device) # (1, 3, 3)
T = -torch.bmm(R.transpose(1, 2), self.camera_position[None, :, None])[:, :, 0] # (1, 3)
image = self.renderer(meshes_world=self.meshes.clone(), R=R, T=T)
# Calculate the silhouette loss
loss = torch.sum((image[..., 3] - self.image_ref) ** 2)
return loss, image
Now we can create an instance of the model above and set up an optimizer for the camera position parameter.
# We will save images periodically and compose them into a GIF.
filename_output = "./teapot_optimization_demo.gif"
writer = imageio.get_writer(filename_output, mode='I', duration=0.3)
# Initialize a model using the renderer, mesh and reference image
model = Model(meshes=teapot_mesh, renderer=silhouette_renderer, image_ref=image_ref).to(device)
# Create an optimizer. Here we are using Adam and we pass in the parameters of the model
optimizer = torch.optim.Adam(model.parameters(), lr=0.05)
plt.figure(figsize=(10, 10))
_, image_init = model()
plt.subplot(1, 2, 1)
plt.imshow(image_init.detach().squeeze().cpu().numpy()[..., 3])
plt.grid(False)
plt.title("Starting position")
plt.subplot(1, 2, 2)
plt.imshow(model.image_ref.cpu().numpy().squeeze())
plt.grid(False)
plt.title("Reference silhouette");
We run several iterations of the forward and backward pass and save outputs every 10 iterations. When this has finished take a look at ./teapot_optimization_demo.gif for a cool gif of the optimization process!
loop = tqdm(range(200))
for i in loop:
optimizer.zero_grad()
loss, _ = model()
loss.backward()
optimizer.step()
loop.set_description('Optimizing (loss %.4f)' % loss.data)
if loss.item() < 200:
break
# Save outputs to create a GIF.
if i % 10 == 0:
R = look_at_rotation(model.camera_position[None, :], device=model.device)
T = -torch.bmm(R.transpose(1, 2), model.camera_position[None, :, None])[:, :, 0] # (1, 3)
image = phong_renderer(meshes_world=model.meshes.clone(), R=R, T=T)
image = image[0, ..., :3].detach().squeeze().cpu().numpy()
image = img_as_ubyte(image)
writer.append_data(image)
plt.figure()
plt.imshow(image[..., :3])
plt.title("iter: %d, loss: %0.2f" % (i, loss.data))
plt.axis("off")
writer.close()
In this tutorial we learnt how to load a mesh from an obj file, initialize a PyTorch3D datastructure called Meshes, set up an Renderer consisting of a Rasterizer and a Shader, set up an optimization loop including a Model and a loss function, and run the optimization.