Slicing

Ipyvolume has no special support for slicing 3d volumes, but has the options to support this. To understand how to do this, we will create a 3d plot, with a plane, which will be controlled using the mouse, and later on texture map this with the same data as the volumetric data.

Simple guassian blob

Lets start with a simple scatter plot.

import ipyvolume as ipv
fig = ipv.figure()
scatter = ipv.examples.gaussian(show=False)
ipv.show()

Now, we add a plane at z=0.

plane = ipv.plot_plane("z");

By holding the shift key and hovering the mouse at the edges of the bounding box (or activate slice mode in the toolbar, and click), we modify the slice_z property. By linking the slice_z property to the z_offset property of the mesh/plane, we can interactively move the plane. Note that in order to change the z_offset, you need to hover the mouse at the sides of the bounding box, which means you need to make sides of the bounding box visible.

import ipywidgets as widgets
widgets.jslink((fig, 'slice_z'), (plane, 'z_offset'));

Adding a texture

This plane can be texture mapped with additional information, for instance, a heatmap. We use vaex with maplotlib to create a simple 2d heatmap PIL Image.

## Uncomment to try
# import vaex
# import matplotlib.pylab as plt
# import PIL.Image

# df = vaex.from_arrays(x=scatter.x, y=scatter.y)

# fig2d = plt.figure()
# ax = fig2d.add_axes([0, 0, 1, 1])
# df.viz.heatmap(df.x, df.y, shape=64, show=False, colorbar=False, tight_layout=False)
# fig2d.axes[0].axis('off');
# plt.draw()
# image = PIL.Image.frombytes('RGB', fig2d.canvas.get_width_height(), fig2d.canvas.tostring_rgb())
# plt.close()
# image

On just download an image:

# example how put a png as texture
import PIL.Image
import requests
import io

url = 'https://vaex.io/img/logos/spiral-small.png'
r = requests.get(url, stream=True)
f = io.BytesIO(r.content)
image = PIL.Image.open(f)

And assign it to the plane’s texture. Note that we should also set its u and v coordinates, so we know where the edges of the texture map should go:

plane.u = [0.0, 1.0, 1.0, 0.0]
plane.v = [0.0, 0.0, 1.0, 1.0]
plane.texture = image

Slicing a volume

We can also, texture map a mesh (a plane is a mesh) with a 3d texture, from the volumetric data.

import ipyvolume as ipv
fig = ipv.figure()
volume = ipv.examples.head(show=False, description="Patient X")
ipv.show()

Downloading https://github.com/maartenbreddels/ipyvolume/raw/master/datasets/male.raw.bz2 to /home/docs/.ipyvolume/datasets/male.raw.bz2

--2021-11-01 12:38:53--  https://github.com/maartenbreddels/ipyvolume/raw/master/datasets/male.raw.bz2
Resolving github.com (github.com)... 140.82.114.3
Connecting to github.com (github.com)|140.82.114.3|:443... connected.
HTTP request sent, awaiting response... 302 Found
Location: https://raw.githubusercontent.com/maartenbreddels/ipyvolume/master/datasets/male.raw.bz2 [following]
--2021-11-01 12:38:53--  https://raw.githubusercontent.com/maartenbreddels/ipyvolume/master/datasets/male.raw.bz2
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.109.133, 185.199.108.133, 185.199.111.133, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.109.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 2305035 (2.2M) [application/octet-stream]
Saving to: ‘/home/docs/.ipyvolume/datasets/male.raw.bz2’

male.raw.bz2        100%[===================>]   2.20M  --.-KB/s    in 0.08s

2021-11-01 12:38:53 (29.3 MB/s) - ‘/home/docs/.ipyvolume/datasets/male.raw.bz2’ saved [2305035/2305035]

We now add 3 planes, and pass our volume so it can be used as a texture map.

slice_x = ipv.plot_plane('x', volume=volume, description="Slice X", description_color="black", icon="mdi-knife")
slice_y = ipv.plot_plane('y', volume=volume, description="Slice Y", description_color="black", icon="mdi-knife")
slice_z = ipv.plot_plane('z', volume=volume, description="Slice Z", description_color="black", icon="mdi-knife",
                         visible=False)

Again, by connecting the slice coordinates to the offsets of the planes, we can create 3 slicing planes that can be controlled interactively.

import ipywidgets as widgets
widgets.jslink((fig, 'slice_x'), (slice_x, 'x_offset'))
widgets.jslink((fig, 'slice_y'), (slice_y, 'y_offset'))
widgets.jslink((fig, 'slice_z'), (slice_z, 'z_offset'));

Note that you can save the output to an html file, and the slicing will still work without a connected kernel.

# uncomment to save
ipv.save("slice.html", devmode=True)

screencapture