Lookup Tables

Demonstrate the usage of a lookup table within PyVista

The pyvista.LookupTable can be used to have fine-tuned control over the mapping between a pyvista.DataSet’s scalars and RGBA colors.

from __future__ import annotations

import pyvista as pv
from pyvista import examples


# download an example dataset
bracket = examples.download_fea_bracket().cell_data_to_point_data()
bracket

Header	Data Arrays
UnstructuredGrid Information N Cells 56786 N Points 102578 X Bounds -2.000e+00, 2.000e+00 Y Bounds -2.000e+00, 2.000e+00 Z Bounds 0.000e+00, 3.250e+00 N Arrays 1	Name Field Type N Comp Min Max Equivalent (von-Mises) Stress (psi) Points float32 1 1.437e-02 2.327e+01

Default Color Map - Lookup Table

First, let’s plot using the default color map, “viridis”. Internally, PyVista will automatically create a lookup table to map the scalars (stored here within point_data) to RGBA colors. This is shown here as a nested attribute to the pyvista.DataSetMapper and it has a helpful repr method:

pl = pv.Plotter()
actor = pl.add_mesh(bracket)
actor.mapper.lookup_table

LookupTable (0x7f8b12418fa0)
  Table Range:                (0.014368999749422073, 23.26799964904785)
  N Values:                   256
  Above Range Color:          None
  Below Range Color:          None
  NAN Color:                  Color(name='darkgray', hex='#a9a9a9ff', opacity=255)
  Log Scale:                  False
  Color Map:                  "viridis"

Plot the Lookup Table

You can also plot lookup table to see the mapping between the scalar values (here, between 0 and 23.3) and RGBA colors.

pl = pv.Plotter()
actor = pl.add_mesh(bracket)
actor.mapper.lookup_table.plot()

Plot the DataSet

Let’s plot the dataset using the automatically generated lookup table.

pl = pv.Plotter()
pl.add_mesh(bracket)
pl.show()

Create a Custom Lookup Table using a Matplotlib Color Map

Here we create a lookup table with a narrow table range (same as clim) and color values above and below the range.

lut = pv.LookupTable(cmap='magma')
lut.scalar_range = (5, 15)
lut.below_range_color = pv.Color('grey', opacity=0.5)
lut.above_range_color = 'r'
lut.plot()

Plot the bracket with the custom colormap

You can set assign the lookup table when using add_mesh with cmap=.

pl = pv.Plotter()
actor = pl.add_mesh(bracket, cmap=lut, lighting=False)
pl.show()

Create a Custom Lookup Table using VTK’s Methods

If you want to create a completely unique color map, you can use attributes like pyvista.LookupTable.hue_range and pyvista.LookupTable.value_range to create your own lookup table.

lut = pv.LookupTable()
lut.value_range = (0.35, 1)  # dark grey to white
lut.hue_range = (0.35, 0.7)  # green to cyna
lut.saturation_range = (0.75, 0.5)  # reduce saturation near the upper end
lut.alpha_range = (0.0, 0.9)  #
lut.scalar_range = (2, 18)
lut.plot()

Plot the bracket with the custom colormap

Assign this custom color map to the plotter and disable lighting to improve the plot.

pl = pv.Plotter()
actor = pl.add_mesh(bracket, cmap=lut, lighting=False)
pl.show()

Custom colormap with widgets

Here we plot the scalars and dynamically change the lookup table through widgets. We create several overlapping single slider widgets to simulate a double ended slider widget.

This example just controls the alpha channel.

pl = pv.Plotter()
actor = pl.add_mesh(bracket, cmap=lut, lighting=False)
pl.add_text('Alpha Range Demo')


def set_min_alpha(min_value):
    max_value = lut.alpha_range[1]
    if min_value > max_value:
        # force the movement of the maximum value
        max_value = min_value
        pl.slider_widgets[1].GetRepresentation().SetValue(max_value)
    lut.alpha_range = (min_value, max_value)


def set_max_alpha(max_value):
    min_value = lut.alpha_range[0]
    if max_value < min_value:
        # force the movement of the minimum value
        min_value = max_value
        pl.slider_widgets[0].GetRepresentation().SetValue(min_value)

    lut.alpha_range = (min_value, max_value)


# create two overlapping slider bars by hiding the tube of the second
pl.add_slider_widget(
    set_min_alpha,
    (0, 1),
    value=lut.alpha_range[0],
    interaction_event='always',
    title='Alpha Range',
    tube_width=0.003,
)
pl.add_slider_widget(
    set_max_alpha,
    (0, 1),
    value=lut.alpha_range[1],
    interaction_event='always',
    tube_width=0.0,
)

pl.show()

Control Several Lookup Table Attributes

Demonstrate the use of several slider bar widgets with lookup table callbacks.

# Create a new lookup table with oranges
lut = pv.LookupTable()
lut.value_range = (0.3, 0.75)
lut.hue_range = (0.0, 0.095)
lut.saturation_range = (0.0, 0.67)
lut.alpha_range = (0.0, 1.0)
lut.scalar_range = (2, 18)

scalars_rng = (bracket.active_scalars.min(), bracket.active_scalars.max())


def make_double_slider(attr, idx):
    """Create a double slider for a given lookup table attribute."""

    def set_min(min_value):
        max_value = getattr(lut, attr)[1]
        if min_value > max_value:
            # force the movement of the maximum value
            max_value = min_value
            pl.slider_widgets[idx * 2 + 1].GetRepresentation().SetValue(max_value)
        setattr(lut, attr, (min_value, max_value))

        if attr == 'scalar_range':
            actor.mapper.scalar_range = getattr(lut, attr)

    def set_max(max_value):
        min_value = getattr(lut, attr)[0]
        if max_value < min_value:
            # force the movement of the minimum value
            min_value = max_value
            pl.slider_widgets[idx * 2].GetRepresentation().SetValue(min_value)
        setattr(lut, attr, (min_value, max_value))

        if attr == 'scalar_range':
            actor.mapper.scalar_range = getattr(lut, attr)

    rng = scalars_rng if attr == 'scalar_range' else (0, 1)

    # create two overlapping slider bars by hiding the tube of the second
    pl.add_slider_widget(
        set_min,
        rng,
        value=getattr(lut, attr)[0],
        interaction_event='always',
        title=' '.join(attr.split('_')).capitalize(),
        tube_width=0.003,
        pointa=(0.6, 0.9 - 0.165 * idx),
        pointb=(0.9, 0.9 - 0.165 * idx),
    )
    pl.add_slider_widget(
        set_max,
        rng,
        value=getattr(lut, attr)[1],
        interaction_event='always',
        tube_width=0.0,
        pointa=(0.6, 0.9 - 0.165 * idx),
        pointb=(0.9, 0.9 - 0.165 * idx),
    )


pl = pv.Plotter()
actor = pl.add_mesh(bracket, cmap=lut, lighting=False)
make_double_slider('alpha_range', 0)
make_double_slider('hue_range', 1)
make_double_slider('value_range', 2)
make_double_slider('saturation_range', 3)
make_double_slider('scalar_range', 4)

pl.camera_position = [(9.021, 5.477, 7.780), (-0.679, 1.349, 0.874), (-0.498, -0.228, 0.836)]
cpos = pl.show(return_cpos=True)