vispy 0.14.0__cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl

vispy/visuals/image.py

@@ -0,0 +1,688 @@
+# -*- coding: utf-8 -*-
+# Copyright (c) Vispy Development Team. All Rights Reserved.
+# Distributed under the (new) BSD License. See LICENSE.txt for more info.
+"""Primitive 2D image visual class."""
+
+from __future__ import division
+
+import warnings
+
+import numpy as np
+
+from ..gloo import Texture2D, VertexBuffer
+from ..color import get_colormap
+from .shaders import Function, FunctionChain
+from .transforms import NullTransform
+from .visual import Visual
+from ..io import load_spatial_filters
+from ._scalable_textures import CPUScaledTexture2D, GPUScaledTexture2D
+
+
+_VERTEX_SHADER = """
+uniform int method;  // 0=subdivide, 1=impostor
+attribute vec2 a_position;
+attribute vec2 a_texcoord;
+varying vec2 v_texcoord;
+
+void main() {
+    v_texcoord = a_texcoord;
+    gl_Position = $transform(vec4(a_position, 0., 1.));
+}
+"""
+
+_FRAGMENT_SHADER = """
+uniform vec2 image_size;
+uniform int method;  // 0=subdivide, 1=impostor
+uniform sampler2D u_texture;
+varying vec2 v_texcoord;
+
+vec4 map_local_to_tex(vec4 x) {
+    // Cast ray from 3D viewport to surface of image
+    // (if $transform does not affect z values, then this
+    // can be optimized as simply $transform.map(x) )
+    vec4 p1 = $transform(x);
+    vec4 p2 = $transform(x + vec4(0, 0, 0.5, 0));
+    p1 /= p1.w;
+    p2 /= p2.w;
+    vec4 d = p2 - p1;
+    float f = p2.z / d.z;
+    vec4 p3 = p2 - d * f;
+
+    // finally map local to texture coords
+    return vec4(p3.xy / image_size, 0, 1);
+}
+
+
+void main()
+{
+    vec2 texcoord;
+    if( method == 0 ) {
+        texcoord = v_texcoord;
+    }
+    else {
+        // vertex shader outputs clip coordinates;
+        // fragment shader maps to texture coordinates
+        texcoord = map_local_to_tex(vec4(v_texcoord, 0, 1)).xy;
+    }
+
+    gl_FragColor = $color_transform($get_data(texcoord));
+}
+"""  # noqa
+
+_INTERPOLATION_TEMPLATE = """
+    #include "misc/spatial-filters.frag"
+    vec4 texture_lookup_filtered(vec2 texcoord) {
+        if(texcoord.x < 0.0 || texcoord.x > 1.0 ||
+        texcoord.y < 0.0 || texcoord.y > 1.0) {
+            discard;
+        }
+        return %s($texture, $shape, texcoord);
+    }"""
+
+_TEXTURE_LOOKUP = """
+    vec4 texture_lookup(vec2 texcoord) {
+        if(texcoord.x < 0.0 || texcoord.x > 1.0 ||
+        texcoord.y < 0.0 || texcoord.y > 1.0) {
+            discard;
+        }
+        return texture2D($texture, texcoord);
+    }"""
+
+_APPLY_CLIM_FLOAT = """
+    float apply_clim(float data) {
+        // If data is NaN, don't draw it at all
+        // http://stackoverflow.com/questions/11810158/how-to-deal-with-nan-or-inf-in-opengl-es-2-0-shaders
+        if (!(data <= 0.0 || 0.0 <= data)) {
+            discard;
+        }
+        data = clamp(data, min($clim.x, $clim.y), max($clim.x, $clim.y));
+        data = (data - $clim.x) / ($clim.y - $clim.x);
+        return data;
+    }"""
+
+_APPLY_CLIM = """
+    vec4 apply_clim(vec4 color) {
+        // Handle NaN values
+        // http://stackoverflow.com/questions/11810158/how-to-deal-with-nan-or-inf-in-opengl-es-2-0-shaders
+        color.r = !(color.r <= 0.0 || 0.0 <= color.r) ? min($clim.x, $clim.y) : color.r;
+        color.g = !(color.g <= 0.0 || 0.0 <= color.g) ? min($clim.x, $clim.y) : color.g;
+        color.b = !(color.b <= 0.0 || 0.0 <= color.b) ? min($clim.x, $clim.y) : color.b;
+        color.a = !(color.a <= 0.0 || 0.0 <= color.a) ? 0 : color.a;
+        color.rgb = clamp(color.rgb, min($clim.x, $clim.y), max($clim.x, $clim.y));
+        color.rgb = (color.rgb - $clim.x) / ($clim.y - $clim.x);
+        return max(color, 0.0);
+    }
+"""
+
+_APPLY_GAMMA_FLOAT = """
+    float apply_gamma(float data) {
+        return pow(data, $gamma);
+    }"""
+
+_APPLY_GAMMA = """
+    vec4 apply_gamma(vec4 color) {
+        color.rgb = pow(color.rgb, vec3($gamma));
+        return color;
+    }
+"""
+
+_NULL_COLOR_TRANSFORM = 'vec4 pass(vec4 color) { return color; }'
+
+_C2L_RED = 'float cmap(vec4 color) { return color.r; }'
+
+_CUSTOM_FILTER = """
+vec4 texture_lookup(vec2 texcoord) {
+    // based on https://gist.github.com/kingbedjed/373c8811efcf1b3a155d29a13c1e5b61
+    vec2 tex_pixel = 1 / $shape;
+    vec2 kernel_pixel = 1 / $kernel_shape;
+    vec2 sampling_corner = texcoord - ($kernel_shape / 2 * tex_pixel);
+
+    // loop over kernel pixels
+    vec2 kernel_pos, tex_pos;
+    vec4 color = vec4(0);
+    float weight;
+
+    // offset 0.5 to sample center of pixels
+    for (float i = 0.5; i < $kernel_shape.x; i++) {
+        for (float j = 0.5; j < $kernel_shape.y; j++) {
+            kernel_pos = vec2(i, j) * kernel_pixel;
+            tex_pos = sampling_corner + vec2(i, j) * tex_pixel;
+            // TODO: allow other edge effects, like mirror or wrap
+            if (tex_pos.x >= 0 && tex_pos.y >= 0 && tex_pos.x <= 1 && tex_pos.y <= 1) {
+                weight = texture2D($kernel, kernel_pos).r;
+                // make sure to clamp or we sample outside
+                color += texture2D($texture, clamp(tex_pos, 0, 1)) * weight;
+            }
+        }
+    }
+
+    return color;
+}
+"""
+
+
+class ImageVisual(Visual):
+    """Visual subclass displaying an image.
+
+    Parameters
+    ----------
+    data : ndarray
+        ImageVisual data. Can be shape (M, N), (M, N, 3), or (M, N, 4).
+        If floating point data is provided and contains NaNs, they will
+        be made transparent (discarded) for the single band data case when
+        scaling is done on the GPU (see ``texture_format``). On the CPU,
+        single band NaNs are mapped to 0 as they are sent to the GPU which
+        result in them using the lowest ``clim`` value in the GPU.
+        For RGB data, NaNs will be mapped to the lowest ``clim`` value.
+        If the Alpha band is NaN it will be mapped to 0 (transparent).
+        Note that NaN handling is not required by some OpenGL implementations
+        and NaNs may be treated differently on some systems (ex. as 0s).
+    method : str
+        Selects method of rendering image in case of non-linear transforms.
+        Each method produces similar results, but may trade efficiency
+        and accuracy. If the transform is linear, this parameter is ignored
+        and a single quad is drawn around the area of the image.
+
+            * 'auto': Automatically select 'impostor' if the image is drawn
+              with a nonlinear transform; otherwise select 'subdivide'.
+            * 'subdivide': ImageVisual is represented as a grid of triangles
+              with texture coordinates linearly mapped.
+            * 'impostor': ImageVisual is represented as a quad covering the
+              entire view, with texture coordinates determined by the
+              transform. This produces the best transformation results, but may
+              be slow.
+
+    grid: tuple (rows, cols)
+        If method='subdivide', this tuple determines the number of rows and
+        columns in the image grid.
+    cmap : str | ColorMap
+        Colormap to use for luminance images.
+    clim : str | tuple
+        Limits to use for the colormap. I.e. the values that map to black and white
+        in a gray colormap. Can be 'auto' to auto-set bounds to
+        the min and max of the data. If not given or None, 'auto' is used.
+    gamma : float
+        Gamma to use during colormap lookup.  Final color will be cmap(val**gamma).
+        by default: 1.
+    interpolation : str
+        Selects method of texture interpolation. Makes use of the two hardware
+        interpolation methods and the available interpolation methods defined
+        in vispy/gloo/glsl/misc/spatial_filters.frag
+
+            * 'nearest': Default, uses 'nearest' with Texture interpolation.
+            * 'linear': uses 'linear' with Texture interpolation.
+            * 'hanning', 'hamming', 'hermite', 'kaiser', 'quadric', 'cubic',
+                'catrom', 'mitchell', 'spline16', 'spline36', 'gaussian',
+                'bessel', 'sinc', 'lanczos', 'blackman'
+            * 'custom': uses the sampling kernel provided through 'custom_kernel'.
+    texture_format: numpy.dtype | str | None
+        How to store data on the GPU. OpenGL allows for many different storage
+        formats and schemes for the low-level texture data stored in the GPU.
+        Most common is unsigned integers or floating point numbers.
+        Unsigned integers are the most widely supported while other formats
+        may not be supported on older versions of OpenGL or with older GPUs.
+        Default value is ``None`` which means data will be scaled on the
+        CPU and the result stored in the GPU as an unsigned integer. If a
+        numpy dtype object, an internal texture format will be chosen to
+        support that dtype and data will *not* be scaled on the CPU. Not all
+        dtypes are supported. If a string, then
+        it must be one of the OpenGL internalformat strings described in the
+        table on this page: https://www.khronos.org/registry/OpenGL-Refpages/gl4/html/glTexImage2D.xhtml
+        The name should have `GL_` removed and be lowercase (ex.
+        `GL_R32F` becomes ``'r32f'``). Lastly, this can also be the string
+        ``'auto'`` which will use the data type of the provided image data
+        to determine the internalformat of the texture.
+        When this is specified (not ``None``) data is scaled on the
+        GPU which allows for faster color limit changes. Additionally, when
+        32-bit float data is provided it won't be copied before being
+        transferred to the GPU.
+    custom_kernel: numpy.ndarray
+        Kernel used for texture sampling when interpolation is set to 'custom'.
+    **kwargs : dict
+        Keyword arguments to pass to `Visual`.
+
+    Notes
+    -----
+    The colormap functionality through ``cmap`` and ``clim`` are only used
+    if the data are 2D.
+    """
+
+    _shaders = {
+        'vertex': _VERTEX_SHADER,
+        'fragment': _FRAGMENT_SHADER,
+    }
+
+    _func_templates = {
+        'texture_lookup_interpolated': _INTERPOLATION_TEMPLATE,
+        'texture_lookup_custom': _CUSTOM_FILTER,
+        'texture_lookup': _TEXTURE_LOOKUP,
+        'clim_float': _APPLY_CLIM_FLOAT,
+        'clim': _APPLY_CLIM,
+        'gamma_float': _APPLY_GAMMA_FLOAT,
+        'gamma': _APPLY_GAMMA,
+        'null_color_transform': _NULL_COLOR_TRANSFORM,
+        'red_to_luminance': _C2L_RED,
+    }
+
+    def __init__(self, data=None, method='auto', grid=(1, 1),
+                 cmap='viridis', clim='auto', gamma=1.0,
+                 interpolation='nearest', texture_format=None,
+                 custom_kernel=np.ones((1, 1)), **kwargs):
+        """Initialize image properties, texture storage, and interpolation methods."""
+        self._data = None
+
+        # load 'float packed rgba8' interpolation kernel
+        # to load float interpolation kernel use
+        # `load_spatial_filters(packed=False)`
+        kernel, interpolation_names = load_spatial_filters()
+
+        self._kerneltex = Texture2D(kernel, interpolation='nearest')
+        # The unpacking can be debugged by changing "spatial-filters.frag"
+        # to have the "unpack" function just return the .r component. That
+        # combined with using the below as the _kerneltex allows debugging
+        # of the pipeline
+        # self._kerneltex = Texture2D(kernel, interpolation='linear',
+        #                             internalformat='r32f')
+
+        interpolation_names, interpolation_fun = self._init_interpolation(
+            interpolation_names)
+        self._interpolation_names = interpolation_names
+        self._interpolation_fun = interpolation_fun
+        self._interpolation = interpolation
+        if self._interpolation not in self._interpolation_names:
+            raise ValueError("interpolation must be one of %s" %
+                             ', '.join(self._interpolation_names))
+
+        self._method = method
+        self._grid = grid
+        self._need_texture_upload = True
+        self._need_vertex_update = True
+        self._need_colortransform_update = True
+        self._need_interpolation_update = True
+        self._texture = self._init_texture(data, texture_format)
+        self._subdiv_position = VertexBuffer()
+        self._subdiv_texcoord = VertexBuffer()
+
+        # impostor quad covers entire viewport
+        vertices = np.array([[-1, -1], [1, -1], [1, 1],
+                             [-1, -1], [1, 1], [-1, 1]],
+                            dtype=np.float32)
+        self._impostor_coords = VertexBuffer(vertices)
+        self._null_tr = NullTransform()
+
+        self._init_view(self)
+
+        Visual.__init__(self, vcode=self._shaders['vertex'], fcode=self._shaders['fragment'])
+        self.set_gl_state('translucent', cull_face=False)
+        self._draw_mode = 'triangles'
+
+        # define _data_lookup_fn as None, will be setup in
+        # self._build_interpolation()
+        self._data_lookup_fn = None
+
+        self.clim = clim or "auto"  # None -> "auto"
+        self.cmap = cmap
+        self.gamma = gamma
+        self.custom_kernel = custom_kernel
+
+        if data is not None:
+            self.set_data(data)
+        self.freeze()
+
+    def _init_interpolation(self, interpolation_names):
+        # create interpolation shader functions for available interpolations
+        fun = [Function(self._func_templates['texture_lookup_interpolated'] % (n + '2D'))
+               for n in interpolation_names]
+        interpolation_names = [n.lower() for n in interpolation_names]
+
+        # add custom filter
+        fun.append(Function(self._func_templates['texture_lookup_custom']))
+        interpolation_names.append('custom')
+
+        interpolation_fun = dict(zip(interpolation_names, fun))
+        interpolation_names = tuple(sorted(interpolation_names))
+
+        # overwrite "nearest" and "linear" spatial-filters
+        # with  "hardware" interpolation _data_lookup_fn
+        hardware_lookup = Function(self._func_templates['texture_lookup'])
+        interpolation_fun['nearest'] = hardware_lookup
+        interpolation_fun['linear'] = hardware_lookup
+        # alias bilinear to linear and bicubic to cubic (but deprecate)
+        interpolation_names = interpolation_names + ('bilinear', 'bicubic')
+        return interpolation_names, interpolation_fun
+
+    def _init_texture(self, data, texture_format, **texture_kwargs):
+        if self._interpolation == 'linear':
+            texture_interpolation = 'linear'
+        else:
+            texture_interpolation = 'nearest'
+
+        if texture_format is None:
+            tex = CPUScaledTexture2D(
+                data, interpolation=texture_interpolation,
+                **texture_kwargs
+            )
+        else:
+            tex = GPUScaledTexture2D(
+                data, internalformat=texture_format,
+                interpolation=texture_interpolation,
+                **texture_kwargs
+            )
+        return tex
+
+    def set_data(self, image, copy=False):
+        """Set the image data.
+
+        Parameters
+        ----------
+        image : array-like
+            The image data.
+        texture_format : str or None
+
+        """
+        data = np.array(image, copy=copy)
+        if np.iscomplexobj(data):
+            raise TypeError(
+                "Complex data types not supported. Please use 'ComplexImage' instead"
+            )
+        # can the texture handle this data?
+        self._texture.check_data_format(data)
+        if self._data is None or self._data.shape[:2] != data.shape[:2]:
+            # Only rebuild if the size of the image changed
+            self._need_vertex_update = True
+        self._data = data
+        self._need_texture_upload = True
+
+    def view(self):
+        """Get the :class:`vispy.visuals.visual.VisualView` for this visual."""
+        v = Visual.view(self)
+        self._init_view(v)
+        return v
+
+    def _init_view(self, view):
+        # Store some extra variables per-view
+        view._need_method_update = True
+        view._method_used = None
+
+    @property
+    def clim(self):
+        """Get color limits used when rendering the image (cmin, cmax)."""
+        return self._texture.clim
+
+    @clim.setter
+    def clim(self, clim):
+        if self._texture.set_clim(clim):
+            self._need_texture_upload = True
+        self._update_colortransform_clim()
+        self.update()
+
+    def _update_colortransform_clim(self):
+        if self._need_colortransform_update:
+            # we are going to rebuild anyway so just do it later
+            return
+        try:
+            norm_clims = self._texture.clim_normalized
+        except RuntimeError:
+            return
+        else:
+            # shortcut so we don't have to rebuild the whole color transform
+            self.shared_program.frag['color_transform'][1]['clim'] = norm_clims
+
+    @property
+    def cmap(self):
+        """Get the colormap object applied to luminance (single band) data."""
+        return self._cmap
+
+    @cmap.setter
+    def cmap(self, cmap):
+        self._cmap = get_colormap(cmap)
+        self._need_colortransform_update = True
+        self.update()
+
+    @property
+    def gamma(self):
+        """Get the gamma used when rendering the image."""
+        return self._gamma
+
+    @gamma.setter
+    def gamma(self, value):
+        """Set gamma used when rendering the image."""
+        if value <= 0:
+            raise ValueError("gamma must be > 0")
+        self._gamma = float(value)
+        # shortcut so we don't have to rebuild the color transform
+        if not self._need_colortransform_update:
+            self.shared_program.frag['color_transform'][2]['gamma'] = self._gamma
+        self.update()
+
+    @property
+    def method(self):
+        """Get rendering method name."""
+        return self._method
+
+    @method.setter
+    def method(self, m):
+        if self._method != m:
+            self._method = m
+            self._need_vertex_update = True
+            self.update()
+
+    @property
+    def size(self):
+        """Get size of the image (width, height)."""
+        return self._data.shape[:2][::-1]
+
+    @property
+    def interpolation(self):
+        """Get interpolation algorithm name."""
+        return self._interpolation
+
+    @interpolation.setter
+    def interpolation(self, i):
+        if i not in self._interpolation_names:
+            raise ValueError("interpolation must be one of %s" %
+                             ', '.join(self._interpolation_names))
+        if self._interpolation != i:
+            self._interpolation = i
+            self._need_interpolation_update = True
+            self.update()
+
+    @property
+    def interpolation_functions(self):
+        """Get names of possible interpolation methods."""
+        return self._interpolation_names
+
+    @property
+    def custom_kernel(self):
+        """Kernel used by 'custom' interpolation for texture sampling"""
+        return self._custom_kernel
+
+    @custom_kernel.setter
+    def custom_kernel(self, value):
+        value = np.asarray(value, dtype=np.float32)
+        if value.ndim != 2:
+            raise ValueError(f'kernel must have 2 dimensions; got {value.ndim}')
+        self._custom_kernel = value
+        self._custom_kerneltex = Texture2D(value, interpolation='nearest', internalformat='r32f')
+        if self._data_lookup_fn is not None and 'kernel' in self._data_lookup_fn:
+            self._data_lookup_fn['kernel'] = self._custom_kerneltex
+            self._data_lookup_fn['kernel_shape'] = value.shape[::-1]
+        self.update()
+
+    # The interpolation code could be transferred to a dedicated filter
+    # function in visuals/filters as discussed in #1051
+    def _build_interpolation(self):
+        """Rebuild the _data_lookup_fn for different interpolations."""
+        interpolation = self._interpolation
+        # alias bilinear to linear
+        if interpolation == 'bilinear':
+            warnings.warn(
+                "'bilinear' interpolation is Deprecated. Use 'linear' instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+            interpolation = 'linear'
+        # alias bicubic to cubic
+        if interpolation == 'bicubic':
+            warnings.warn(
+                "'bicubic' interpolation is Deprecated. Use 'cubic' instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
+            interpolation = 'cubic'
+        self._data_lookup_fn = self._interpolation_fun[interpolation]
+        self.shared_program.frag['get_data'] = self._data_lookup_fn
+
+        # only 'linear' and 'custom' use 'linear' texture interpolation
+        if interpolation in ('linear', 'custom'):
+            texture_interpolation = 'linear'
+        else:
+            texture_interpolation = 'nearest'
+
+        # 'nearest' (and also 'linear') doesn't use spatial_filters.frag
+        # so u_kernel and shape setting is skipped
+        if interpolation not in ('nearest', 'linear'):
+            self._data_lookup_fn['shape'] = self._data.shape[:2][::-1]
+            if interpolation == 'custom':
+                self._data_lookup_fn['kernel'] = self._custom_kerneltex
+                self._data_lookup_fn['kernel_shape'] = self._custom_kernel.shape[::-1]
+            else:
+                self.shared_program['u_kernel'] = self._kerneltex
+
+        if self._texture.interpolation != texture_interpolation:
+            self._texture.interpolation = texture_interpolation
+
+        self._data_lookup_fn['texture'] = self._texture
+
+        self._need_interpolation_update = False
+
+    def _build_vertex_data(self):
+        """Rebuild the vertex buffers for the subdivide method."""
+        grid = self._grid
+        w = 1.0 / grid[1]
+        h = 1.0 / grid[0]
+
+        quad = np.array([[0, 0, 0], [w, 0, 0], [w, h, 0],
+                         [0, 0, 0], [w, h, 0], [0, h, 0]],
+                        dtype=np.float32)
+        quads = np.empty((grid[1], grid[0], 6, 3), dtype=np.float32)
+        quads[:] = quad
+
+        mgrid = np.mgrid[0.:grid[1], 0.:grid[0]].transpose(1, 2, 0)
+        mgrid = mgrid[:, :, np.newaxis, :]
+        mgrid[..., 0] *= w
+        mgrid[..., 1] *= h
+
+        quads[..., :2] += mgrid
+        tex_coords = quads.reshape(grid[1]*grid[0]*6, 3)
+        tex_coords = np.ascontiguousarray(tex_coords[:, :2])
+        vertices = tex_coords * self.size
+
+        self._subdiv_position.set_data(vertices.astype('float32'))
+        self._subdiv_texcoord.set_data(tex_coords.astype('float32'))
+        self._need_vertex_update = False
+
+    def _update_method(self, view):
+        """Decide which method to use for *view* and configure it accordingly."""
+        method = self._method
+        if method == 'auto':
+            if view.transforms.get_transform().Linear:
+                method = 'subdivide'
+            else:
+                method = 'impostor'
+        view._method_used = method
+
+        if method == 'subdivide':
+            view.view_program['method'] = 0
+            view.view_program['a_position'] = self._subdiv_position
+            view.view_program['a_texcoord'] = self._subdiv_texcoord
+        elif method == 'impostor':
+            view.view_program['method'] = 1
+            view.view_program['a_position'] = self._impostor_coords
+            view.view_program['a_texcoord'] = self._impostor_coords
+        else:
+            raise ValueError("Unknown image draw method '%s'" % method)
+
+        self.shared_program['image_size'] = self.size
+        view._need_method_update = False
+        self._prepare_transforms(view)
+
+    def _build_texture(self):
+        try:
+            pre_clims = self._texture.clim_normalized
+        except RuntimeError:
+            pre_clims = "auto"
+        pre_internalformat = self._texture.internalformat
+        # copy was already made on `set_data` if requested
+        self._texture.scale_and_set_data(self._data, copy=False)
+        post_clims = self._texture.clim_normalized
+        post_internalformat = self._texture.internalformat
+        # color transform needs rebuilding if the internalformat was changed
+        # new color limits need to be assigned if the normalized clims changed
+        # otherwise, the original color transform should be fine
+        new_if = post_internalformat != pre_internalformat
+        new_cl = post_clims != pre_clims
+        if new_if:
+            self._need_colortransform_update = True
+        elif new_cl and not self._need_colortransform_update:
+            # shortcut so we don't have to rebuild the whole color transform
+            self.shared_program.frag['color_transform'][1]['clim'] = self._texture.clim_normalized
+        self._need_texture_upload = False
+
+    def _compute_bounds(self, axis, view):
+        if axis > 1:
+            return 0, 0
+        else:
+            return 0, self.size[axis]
+
+    def _build_color_transform(self):
+        if self._data.ndim == 2 or self._data.shape[2] == 1:
+            # luminance data
+            fclim = Function(self._func_templates['clim_float'])
+            fgamma = Function(self._func_templates['gamma_float'])
+            # NOTE: red_to_luminance only uses the red component, fancy internalformats
+            #   may need to use the other components or a different function chain
+            fun = FunctionChain(
+                None, [Function(self._func_templates['red_to_luminance']), fclim, fgamma, Function(self.cmap.glsl_map)]
+            )
+        else:
+            # RGB/A image data (no colormap)
+            fclim = Function(self._func_templates['clim'])
+            fgamma = Function(self._func_templates['gamma'])
+            fun = FunctionChain(None, [Function(self._func_templates['null_color_transform']), fclim, fgamma])
+        fclim['clim'] = self._texture.clim_normalized
+        fgamma['gamma'] = self.gamma
+        return fun
+
+    def _prepare_transforms(self, view):
+        trs = view.transforms
+        prg = view.view_program
+        method = view._method_used
+        if method == 'subdivide':
+            prg.vert['transform'] = trs.get_transform()
+            prg.frag['transform'] = self._null_tr
+        else:
+            prg.vert['transform'] = self._null_tr
+            prg.frag['transform'] = trs.get_transform().inverse
+
+    def _prepare_draw(self, view):
+        if self._data is None:
+            return False
+
+        if self._need_interpolation_update:
+            self._build_interpolation()
+
+        if self._need_texture_upload:
+            self._build_texture()
+
+        if self._need_colortransform_update:
+            prg = view.view_program
+            self.shared_program.frag['color_transform'] = self._build_color_transform()
+            self._need_colortransform_update = False
+            prg['texture2D_LUT'] = self.cmap.texture_lut()
+
+        if self._need_vertex_update:
+            self._build_vertex_data()
+
+        if view._need_method_update:
+            self._update_method(view)