def getTexture(color=None, alpha=None, texture_cache=None, diffuseinit=None):
unique_id = ""
if color:
unique_id += str(id(color.sampler.surface.image))
if alpha:
unique_id += '_' + str(id(alpha.sampler.surface.image))
if texture_cache is not None:
if unique_id in texture_cache:
return texture_cache[unique_id]
image_file = ""
if alpha:
im = pilFromData(alpha.sampler.surface.image.data)
gray = None
for i, band in enumerate(im.getbands()):
if band == 'A':
gray = im.split()[i]
if gray is None:
gray = ImageOps.grayscale(im)
if color:
newim = pilFromData(color.sampler.surface.image.data)
if 'A' not in newim.getbands():
newim = newim.convert('RGBA')
else:
if diffuseinit is None:
diffuseinit = (0, 0, 0, 1)
else:
diffuseinit = tuple(int(v * 255) for v in diffuseinit)
newim = Image.new('RGBA', im.size, diffuseinit)
newim.putalpha(gray)
newbuf = StringIO()
newim.save(newbuf, 'PNG')
image_data = newbuf.getvalue()
image_file = 'whatever.png'
else:
image_file = posixpath.basename(color.sampler.surface.image.path)
im = pilFromData(color.sampler.surface.image.data)
if im:
if 'A' in im.getbands():
im = im.convert('RGBA')
else:
im = im.convert('RGB')
newbuf = StringIO()
im.save(newbuf, 'PNG')
image_data = newbuf.getvalue()
else:
image_data = color.sampler.surface.image.data
tex = textureFromData(image_data, image_file)
if texture_cache is not None:
texture_cache[unique_id] = tex
return tex
def getTexture(color=None, alpha=None, texture_cache=None, diffuseinit=None):
unique_id = ""
if color:
unique_id += str(id(color.sampler.surface.image))
if alpha:
unique_id += '_' + str(id(alpha.sampler.surface.image))
if texture_cache is not None:
if unique_id in texture_cache:
return texture_cache[unique_id]
image_file = ""
if alpha:
im = pilFromData(alpha.sampler.surface.image.data)
gray = None
for i, band in enumerate(im.getbands()):
if band == 'A':
gray = im.split()[i]
if gray is None:
gray = ImageOps.grayscale(im)
if color:
newim = pilFromData(color.sampler.surface.image.data)
if 'A' not in newim.getbands():
newim = newim.convert('RGBA')
else:
if diffuseinit is None:
diffuseinit = (0,0,0,1)
else:
diffuseinit = tuple(int(v*255) for v in diffuseinit)
newim = Image.new('RGBA', im.size, diffuseinit)
newim.putalpha(gray)
newbuf = StringIO()
newim.save(newbuf, 'PNG')
image_data = newbuf.getvalue()
image_file = 'whatever.png'
else:
image_file = posixpath.basename(color.sampler.surface.image.path)
im = pilFromData(color.sampler.surface.image.data)
if im:
if 'A' in im.getbands():
im = im.convert('RGBA')
else:
im = im.convert('RGB')
newbuf = StringIO()
im.save(newbuf, 'PNG')
image_data = newbuf.getvalue()
else:
image_data = color.sampler.surface.image.data
tex = textureFromData(image_data, image_file)
if texture_cache is not None:
texture_cache[unique_id] = tex
return tex
def makeAtlases(mesh):
# get a mapping from path to actual image, since theoretically you could have
# the same image file in multiple image nodes
unique_images = {}
image_scales = {}
for cimg in mesh.images:
path = cimg.path
if path not in unique_images:
unique_images[path] = cimg.pilimage
image_scales[path] = (1,1)
# get a mapping from texture coordinates to all of the images they get bound to
tex2img = getTexcoordToImgMapping(mesh)
# don't consider any texcoords that are used more than once with different images
# could probably have an algorithm that takes this into account, but it would
# require some complex groupings. any images referenced have to also not be
# considered for atlasing, as well as any tex coords that reference them
# also filter out any images that are >= 1024 in either dimension
texs_to_delete = []
imgs_to_delete = []
for texset, imgpaths in tex2img.iteritems():
valid_range = False
if len(imgpaths) == 1:
texarray = mesh.geometries[texset.geom_id] \
.primitives[texset.prim_index] \
.texcoordset[texset.texcoordset_index]
width, height = unique_images[imgpaths[0]].size
tile_x = int(numpy.ceil(numpy.max(texarray[:,0])))
tile_y = int(numpy.ceil(numpy.max(texarray[:,1])))
stretched_width = tile_x * width
stretched_height = tile_y * height
#allow tiling of texcoords if the final tiled image is <= MAX_TILING_DIMENSION
if numpy.min(texarray) < 0.0:
valid_range = False
elif stretched_width > MAX_TILING_DIMENSION or stretched_height > MAX_TILING_DIMENSION:
valid_range = False
else:
valid_range = True
if valid_range:
scale_x, scale_y = image_scales[imgpaths[0]]
scale_x = max(scale_x, tile_x)
scale_y = max(scale_y, tile_y)
image_scales[imgpaths[0]] = (scale_x, scale_y)
if len(imgpaths) > 1 or not valid_range:
for imgpath in imgpaths:
if imgpath not in imgs_to_delete:
imgs_to_delete.append(imgpath)
texs_to_delete.append(texset)
for imgpath, pilimg in unique_images.iteritems():
if max(pilimg.size) > MAX_IMAGE_DIMENSION and imgpath not in imgs_to_delete:
imgs_to_delete.append(imgpath)
for imgpath in imgs_to_delete:
for texset, imgpaths in tex2img.iteritems():
if imgpaths[0] == imgpath and texset not in texs_to_delete:
texs_to_delete.append(texset)
del unique_images[imgpath]
for texset in texs_to_delete:
del tex2img[texset]
# now make a mapping between images and the tex coords that have to be
# updated if it is atlased
img2texs = {}
for imgpath in unique_images:
img2texs[imgpath] = []
for texset, imgpaths in tex2img.iteritems():
if imgpaths[0] == imgpath:
img2texs[imgpath].append(texset)
for path, pilimg in unique_images.iteritems():
tile_x, tile_y = image_scales[path]
width, height = pilimg.size
if tile_x > 1 or tile_y > 1:
if 'A' in pilimg.getbands():
imgformat = 'RGBA'
initval = (0,0,0,255)
else:
imgformat = 'RGB'
initval = (0,0,0)
tiled_img = Image.new(imgformat, (width*tile_x, height*tile_y), initval)
for x in range(tile_x):
for y in range(tile_y):
tiled_img.paste(pilimg, (x*width,y*height))
pilimg = tiled_img
width, height = pilimg.size
#round down to power of 2
width = int(math.pow(2, int(math.log(width, 2))))
height = int(math.pow(2, int(math.log(height, 2))))
if (width, height) != pilimg.size:
pilimg = pilimg.resize((width, height), Image.ANTIALIAS)
unique_images[path] = pilimg
group1, group2 = splitAlphas(unique_images)
to_del = combinePacks(packImages(mesh, img2texs, group1, image_scales),
packImages(mesh, img2texs, group2, image_scales))
if to_del is not None:
for geom, primindices in to_del.iteritems():
for i in sorted(primindices, reverse=True):
del geom.primitives[i]
def packImages(mesh, img2texs, unique_images, image_scales):
#if there aren't at least two images left, nothing to do
if len(unique_images) < 2:
return
#okay, now we can start packing!
rp = RectPack(MAX_IMAGE_DIMENSION, MAX_IMAGE_DIMENSION)
for path, pilimg in unique_images.iteritems():
width, height = pilimg.size
rp.addRectangle(path, width, height)
success = rp.pack()
if not success:
if len(rp.rejects) == len(unique_images):
#this means that nothing could be packed into the max size
# if not a single image can be packed into the max size
# then there's no point in continuing
return
group1 = dict(( (path, pilimg) for path, pilimg in unique_images.iteritems() if path in rp.rejects ))
group2 = dict(( (path, pilimg) for path, pilimg in unique_images.iteritems() if path not in rp.rejects ))
return combinePacks(packImages(mesh, img2texs, group1, image_scales),
packImages(mesh, img2texs, group2, image_scales))
width = rp.width
height = rp.height
print "actually making atlas of size %dx%d with %d subimages referenced by %d texcoords" % \
(width, height, len(unique_images), sum([len(img2texs[imgpath]) for imgpath in unique_images]))
atlasimg = Image.new('RGBA', (width, height), (0,0,0,255))
to_del = {}
for path, pilimg in unique_images.iteritems():
x,y,w,h = rp.getPlacement(path)
atlasimg.paste(pilimg, (x,y,x+w,y+h))
x,y,w,h,width,height = (float(i) for i in (x,y,w,h,width,height))
for texset in img2texs[path]:
geom = mesh.geometries[texset.geom_id]
prim = geom.primitives[texset.prim_index]
texarray = numpy.copy(prim.texcoordset[texset.texcoordset_index])
tile_x, tile_y = (float(i) for i in image_scales[path])
#this shrinks the texcoords to 0,1 range for a tiled image
if tile_x > 1.0:
texarray[:,0] = texarray[:,0] / tile_x
if tile_y > 1.0:
texarray[:,1] = texarray[:,1] / tile_y
#this computes the coordinates of the lowest and highest texel
# if the texcoords go outside that range, rescale so they are inside
# suggestion by nvidia texture atlasing white paper
minx, maxx = numpy.min(texarray[:,0]), numpy.max(texarray[:,0])
miny, maxy = numpy.min(texarray[:,1]), numpy.max(texarray[:,1])
lowest_x = 0.5 / w
lowest_y = 0.5 / h
highest_x = 1.0 - lowest_x
highest_y = 1.0 - lowest_y
if minx < lowest_x or maxx > highest_x:
texarray[:,0] = texarray[:,0] * (highest_x - lowest_x) + lowest_x
if miny < lowest_y or maxy > highest_y:
texarray[:,1] = texarray[:,1] * (highest_y - lowest_y) + lowest_y
#this rescales the texcoords to map to the new atlas location
texarray[:,0] = texarray[:,0] * (w / width) + (x / (width-1))
texarray[:,1] = texarray[:,1] * (h / height) + (1.0 - (y+h)/height)
oldsources = prim.getInputList().getList()
newsources = collada.source.InputList()
for (offset, semantic, source, setid) in oldsources:
if semantic == 'TEXCOORD' and (setid is None or int(setid) == texset.texcoordset_index):
orig_source = source
i=0
while source[1:] in geom.sourceById:
source = orig_source + '-atlas-' + str(i)
i += 1
new_tex_src = collada.source.FloatSource(source[1:], texarray, ('S', 'T'))
geom.sourceById[source[1:]] = new_tex_src
newsources.addInput(offset, semantic, source, setid)
if geom not in to_del:
to_del[geom] = []
to_del[geom].append(texset.prim_index)
if type(prim) is collada.triangleset.TriangleSet:
prim.index.shape = -1
newprim = geom.createTriangleSet(prim.index, newsources, prim.material)
elif type(prim) is collada.polylist.Polylist:
prim.index.shape = -1
prim.vcounts.shape = -1
newprim = geom.createPolylist(prim.index, prim.vcounts, newsources, prim.material)
elif type(prim) is collada.polygons.Polygons:
prim.index.shape = -1
newprim = geom.createPolygons(prim.index, newsources, prim.material)
elif type(prim) is collada.lineset.LineSet:
prim.index.shape = -1
newprim = geom.createLineSet(prim.index, newsources, prim.material)
else:
raise Exception("Unknown primitive type")
geom.primitives.append(newprim)
imgs_deleted = [cimg for cimg in mesh.images if cimg.path in unique_images]
mesh.images = [cimg for cimg in mesh.images if cimg.path not in unique_images]
baseimgid = imgs_deleted[0].id + '-atlas'
baseimgpath = './atlas'
newimgid = baseimgid
newimgpath = baseimgpath
ct = 0
while newimgid in mesh.images or newimgpath + '.png' in [cimg.path for cimg in mesh.images]:
newimgid = baseimgid + '-' + str(ct)
newimgpath = baseimgpath + '-' + str(ct)
ct += 1
newimgpath = newimgpath + '.png'
newcimage = collada.material.CImage(newimgid, newimgpath, mesh)
strbuf = StringIO()
atlasimg.save(strbuf, 'PNG', optimize=True)
newcimage._data = strbuf.getvalue()
mesh.images.append(newcimage)
for effect in mesh.effects:
for param in effect.params:
if type(param) is collada.material.Surface:
if param.image in imgs_deleted:
param.image = newcimage
return to_del
def makeAtlases(mesh):
# get a mapping from path to actual image, since theoretically you could have
# the same image file in multiple image nodes
unique_images = {}
image_scales = {}
for cimg in mesh.images:
path = cimg.path
if path not in unique_images:
unique_images[path] = cimg.pilimage
image_scales[path] = (1, 1)
# get a mapping from texture coordinates to all of the images they get bound to
tex2img = getTexcoordToImgMapping(mesh)
# don't consider any texcoords that are used more than once with different images
# could probably have an algorithm that takes this into account, but it would
# require some complex groupings. any images referenced have to also not be
# considered for atlasing, as well as any tex coords that reference them
# also filter out any images that are >= 1024 in either dimension
texs_to_delete = []
imgs_to_delete = []
for texset, imgpaths in tex2img.iteritems():
valid_range = False
if len(imgpaths) == 1:
texarray = mesh.geometries[texset.geom_id] \
.primitives[texset.prim_index] \
.texcoordset[texset.texcoordset_index]
width, height = unique_images[imgpaths[0]].size
tile_x = int(numpy.ceil(numpy.max(texarray[:, 0])))
tile_y = int(numpy.ceil(numpy.max(texarray[:, 1])))
stretched_width = tile_x * width
stretched_height = tile_y * height
#allow tiling of texcoords if the final tiled image is <= MAX_TILING_DIMENSION
if numpy.min(texarray) < 0.0:
valid_range = False
elif stretched_width > MAX_TILING_DIMENSION or stretched_height > MAX_TILING_DIMENSION:
valid_range = False
else:
valid_range = True
if valid_range:
scale_x, scale_y = image_scales[imgpaths[0]]
scale_x = max(scale_x, tile_x)
scale_y = max(scale_y, tile_y)
image_scales[imgpaths[0]] = (scale_x, scale_y)
if len(imgpaths) > 1 or not valid_range:
for imgpath in imgpaths:
if imgpath not in imgs_to_delete:
imgs_to_delete.append(imgpath)
texs_to_delete.append(texset)
for imgpath, pilimg in unique_images.iteritems():
if max(pilimg.size
) > MAX_IMAGE_DIMENSION and imgpath not in imgs_to_delete:
imgs_to_delete.append(imgpath)
for imgpath in imgs_to_delete:
for texset, imgpaths in tex2img.iteritems():
if imgpaths[0] == imgpath and texset not in texs_to_delete:
texs_to_delete.append(texset)
del unique_images[imgpath]
for texset in texs_to_delete:
del tex2img[texset]
# now make a mapping between images and the tex coords that have to be
# updated if it is atlased
img2texs = {}
for imgpath in unique_images:
img2texs[imgpath] = []
for texset, imgpaths in tex2img.iteritems():
if imgpaths[0] == imgpath:
img2texs[imgpath].append(texset)
for path, pilimg in unique_images.iteritems():
tile_x, tile_y = image_scales[path]
width, height = pilimg.size
if tile_x > 1 or tile_y > 1:
if 'A' in pilimg.getbands():
imgformat = 'RGBA'
initval = (0, 0, 0, 255)
else:
imgformat = 'RGB'
initval = (0, 0, 0)
tiled_img = Image.new(imgformat, (width * tile_x, height * tile_y),
initval)
for x in range(tile_x):
for y in range(tile_y):
tiled_img.paste(pilimg, (x * width, y * height))
pilimg = tiled_img
width, height = pilimg.size
#round down to power of 2
width = int(math.pow(2, int(math.log(width, 2))))
height = int(math.pow(2, int(math.log(height, 2))))
if (width, height) != pilimg.size:
pilimg = pilimg.resize((width, height), Image.ANTIALIAS)
unique_images[path] = pilimg
group1, group2 = splitAlphas(unique_images)
to_del = combinePacks(packImages(mesh, img2texs, group1, image_scales),
packImages(mesh, img2texs, group2, image_scales))
if to_del is not None:
for geom, primindices in to_del.iteritems():
for i in sorted(primindices, reverse=True):
del geom.primitives[i]
def packImages(mesh, img2texs, unique_images, image_scales):
#if there aren't at least two images left, nothing to do
if len(unique_images) < 2:
return
#okay, now we can start packing!
rp = RectPack(MAX_IMAGE_DIMENSION, MAX_IMAGE_DIMENSION)
for path, pilimg in unique_images.iteritems():
width, height = pilimg.size
rp.addRectangle(path, width, height)
success = rp.pack()
if not success:
if len(rp.rejects) == len(unique_images):
#this means that nothing could be packed into the max size
# if not a single image can be packed into the max size
# then there's no point in continuing
return
group1 = dict(((path, pilimg)
for path, pilimg in unique_images.iteritems()
if path in rp.rejects))
group2 = dict(((path, pilimg)
for path, pilimg in unique_images.iteritems()
if path not in rp.rejects))
return combinePacks(packImages(mesh, img2texs, group1, image_scales),
packImages(mesh, img2texs, group2, image_scales))
width = rp.width
height = rp.height
print "actually making atlas of size %dx%d with %d subimages referenced by %d texcoords" % \
(width, height, len(unique_images), sum([len(img2texs[imgpath]) for imgpath in unique_images]))
atlasimg = Image.new('RGBA', (width, height), (0, 0, 0, 255))
to_del = {}
for path, pilimg in unique_images.iteritems():
x, y, w, h = rp.getPlacement(path)
atlasimg.paste(pilimg, (x, y, x + w, y + h))
x, y, w, h, width, height = (float(i)
for i in (x, y, w, h, width, height))
for texset in img2texs[path]:
geom = mesh.geometries[texset.geom_id]
prim = geom.primitives[texset.prim_index]
texarray = numpy.copy(prim.texcoordset[texset.texcoordset_index])
tile_x, tile_y = (float(i) for i in image_scales[path])
#this shrinks the texcoords to 0,1 range for a tiled image
if tile_x > 1.0:
texarray[:, 0] = texarray[:, 0] / tile_x
if tile_y > 1.0:
texarray[:, 1] = texarray[:, 1] / tile_y
#this computes the coordinates of the lowest and highest texel
# if the texcoords go outside that range, rescale so they are inside
# suggestion by nvidia texture atlasing white paper
minx, maxx = numpy.min(texarray[:, 0]), numpy.max(texarray[:, 0])
miny, maxy = numpy.min(texarray[:, 1]), numpy.max(texarray[:, 1])
lowest_x = 0.5 / w
lowest_y = 0.5 / h
highest_x = 1.0 - lowest_x
highest_y = 1.0 - lowest_y
if minx < lowest_x or maxx > highest_x:
texarray[:,
0] = texarray[:,
0] * (highest_x - lowest_x) + lowest_x
if miny < lowest_y or maxy > highest_y:
texarray[:,
1] = texarray[:,
1] * (highest_y - lowest_y) + lowest_y
#this rescales the texcoords to map to the new atlas location
texarray[:, 0] = texarray[:, 0] * (w / width) + (x / (width - 1))
texarray[:, 1] = texarray[:, 1] * (h / height) + (1.0 -
(y + h) / height)
oldsources = prim.getInputList().getList()
newsources = collada.source.InputList()
for (offset, semantic, source, setid) in oldsources:
if semantic == 'TEXCOORD' and (setid is None or int(setid)
== texset.texcoordset_index):
orig_source = source
i = 0
while source[1:] in geom.sourceById:
source = orig_source + '-atlas-' + str(i)
i += 1
new_tex_src = collada.source.FloatSource(
source[1:], texarray, ('S', 'T'))
geom.sourceById[source[1:]] = new_tex_src
newsources.addInput(offset, semantic, source, setid)
if geom not in to_del:
to_del[geom] = []
to_del[geom].append(texset.prim_index)
if type(prim) is collada.triangleset.TriangleSet:
prim.index.shape = -1
newprim = geom.createTriangleSet(prim.index, newsources,
prim.material)
elif type(prim) is collada.polylist.Polylist:
prim.index.shape = -1
prim.vcounts.shape = -1
newprim = geom.createPolylist(prim.index, prim.vcounts,
newsources, prim.material)
elif type(prim) is collada.polygons.Polygons:
prim.index.shape = -1
newprim = geom.createPolygons(prim.index, newsources,
prim.material)
elif type(prim) is collada.lineset.LineSet:
prim.index.shape = -1
newprim = geom.createLineSet(prim.index, newsources,
prim.material)
else:
raise Exception("Unknown primitive type")
geom.primitives.append(newprim)
imgs_deleted = [cimg for cimg in mesh.images if cimg.path in unique_images]
mesh.images = [
cimg for cimg in mesh.images if cimg.path not in unique_images
]
baseimgid = imgs_deleted[0].id + '-atlas'
baseimgpath = './atlas'
newimgid = baseimgid
newimgpath = baseimgpath
ct = 0
while newimgid in mesh.images or newimgpath + '.png' in [
cimg.path for cimg in mesh.images
]:
newimgid = baseimgid + '-' + str(ct)
newimgpath = baseimgpath + '-' + str(ct)
ct += 1
newimgpath = newimgpath + '.png'
newcimage = collada.material.CImage(newimgid, newimgpath, mesh)
strbuf = StringIO()
atlasimg.save(strbuf, 'PNG', optimize=True)
newcimage._data = strbuf.getvalue()
mesh.images.append(newcimage)
for effect in mesh.effects:
for param in effect.params:
if type(param) is collada.material.Surface:
if param.image in imgs_deleted:
param.image = newcimage
return to_del
