Note
Go to the end to download the full example code
Ambient occlusion¶
This example demonstrates the ambient occlusion map effects for MeshBasicMaterial, MeshPhongMaterial, and MeshStandardMaterial.
Note
To run this example, you need a model from the source repo’s example folder. If you are running this example from a local copy of the code (dev install) no further actions are needed. Otherwise, you may have to replace the path below to point to the location of the model.
import os
from pathlib import Path
try:
# modify this line if your model is located elsewhere
model_dir = Path(__file__).parents[1] / "data"
except NameError:
# compatibility with sphinx-gallery
model_dir = Path(os.getcwd()).parent / "data"
Once the path is set correctly, you can use the model as follows:
import imageio.v3 as iio
import numpy as np
from wgpu.gui.auto import WgpuCanvas, run
import pygfx as gfx
# Init
canvas = WgpuCanvas(size=(1200, 400), title="aomap")
renderer = gfx.renderers.WgpuRenderer(canvas)
meshes = gfx.load_mesh(model_dir / "lightmap" / "scene.gltf")
ao_map = iio.imread(model_dir / "lightmap" / "lightmap-ao-shadow.png")
ao_map_tex = gfx.Texture(ao_map, dim=2)
texcoords1 = np.ascontiguousarray(
np.loadtxt(model_dir / "lightmap" / "texcoords1.txt"), dtype="f4"
)
texcoords1 = gfx.Buffer(texcoords1)
# Create camera and controller
camera = gfx.PerspectiveCamera(45, 1)
camera.local.position = 500, 1000, 1500
camera.look_at((0, 0, 0))
controller = gfx.OrbitController(camera, register_events=renderer)
# Text
text_scene = gfx.Scene()
text_camera = gfx.OrthographicCamera(12, 4)
def create_scene(material, x_pos):
scene = gfx.Scene()
m = meshes[0]
m.geometry.texcoords1 = texcoords1
material.ao_map = ao_map_tex
mesh = gfx.Mesh(m.geometry, material)
mesh.local.matrix = m.local.matrix
scene.add(mesh)
# illumination the scene for MeshPhongMaterial and MeshStandardMaterial
scene.add(gfx.AmbientLight(intensity=1.0))
t = gfx.Text(
gfx.TextGeometry(material.__class__.__name__, screen_space=True, font_size=20),
gfx.TextMaterial(),
)
t.local.position = (x_pos, 1.5, 0)
text_scene.add(t)
return scene
vp1 = gfx.Viewport(renderer, (5, 0, 390, 400))
scene1 = create_scene(gfx.MeshBasicMaterial(), -4)
vp2 = gfx.Viewport(renderer, (405, 0, 390, 400))
scene2 = create_scene(gfx.MeshPhongMaterial(), 0)
vp3 = gfx.Viewport(renderer, (805, 0, 390, 400))
scene3 = create_scene(gfx.MeshStandardMaterial(), 4)
vp4 = gfx.Viewport(renderer, (0, 0, 1200, 400))
def animate():
vp1.render(scene1, camera)
vp2.render(scene2, camera)
vp3.render(scene3, camera)
vp4.render(text_scene, text_camera)
renderer.flush()
if __name__ == "__main__":
renderer.request_draw(animate)
run()
Total running time of the script: (0 minutes 0.916 seconds)