creates a collage of images imagei.


creates a collage of images imagei based on their corresponding weights wi.


also creates a collage of images imagei based on their corresponding weights wi.


also creates a collage of images imagei based on their corresponding weights wi.


also creates a collage of images imagei based on their corresponding weights wi.


uses the fitting method to create the collage.


creates a collage of the specified size.

Details and Options

  • ImageCollage creates a collage of a list of images and can be used to create a compact visualization of photographs, graphics, etc.
  • ImageCollage[w->{image1,image2,}] assumes the same weight w for all imagei.
  • Weights wi can be either of one of the following forms:
  • wfinal area proportional to w
    Scaled[s]final area proportional to s multiplied by the original area
  • Possible values for fitting include:
  • "Fill"fit the smaller dimension; crop the other if necessary
    "Fit"fit the whole image; keep the aspect ratio; pad if necessary
    "Stretch"stretch the image to fit by resampling
  • The following options can be specified:
  • BackgroundGrayLevel[0.2]background color
    ImagePaddingNonepadding size to use around each image
    ImageSizeAutomaticdisplay size of the collage
    MethodAutomaticmethod to use for packing images
    PaddingAutomaticpadding scheme to use around each image
    ResamplingAutomaticresampling method
  • Possible settings for ImagePadding are:
  • Noneno padding
    mthe same padding on all sides
    {m,n}m pixels on left and right, n pixels on bottom and top
    {{left,right},{bottom,top}}different padding on different sides
  • Method can take one of the following settings:
  • "Rows"arrange in horizontal layers
    "Columns"arrange in vertical layers
    "Grid"a regular grid of images
    "ClosestPacking"closest packing from the top left


open allclose all

Basic Examples  (3)

Create an image collage from a list of images:

Create a collage from a weighted list of images:

Collage of flags of G7 countries, weighted by their population:

Scope  (4)

Weights  (2)

With no weights specified, almost-equal weights are assumed:

Images will appear to have roughly the same area:

Absolute weights represent a relationship between image areas in the collage:

Scaled weights get multiplied with original image dimensions:

Use Scaled[1] as the weight for all images to preserve original image area ratios:

Compatible Quantity objects can also be used as weights:

Fitting  (1)

Use different fitting schemes to specify how to fit images in the allocated space:

Size  (1)

If no size is specified, the collage size is influenced by the original dimensions as well as the method:

Specify the output image width:

Specify the width and height of the output image:

Create a portrait collage:

Options  (12)

Background  (4)

By default, the background is filled with GrayLevel[0.2]:

Specify a background color:

Use Background->Transparent to create a collage with transparent background:

With Padding->Automatic, images are padded with the background color:

Use a different padding value:

Method  (5)

By default, the packing method is automatically selected:

Specify the packing method:

The "ClosestPacking" method tries to find a packing while preserving relative weighted image areas:

The "Grid" method places the images in a regular grid:

The "Rows" method places images in horizontal layers:

Create vertical layers using Method->"Columns":

With Method->"Grid", by default all cells have the same size, equivalent to ItemSize->All:

Use ItemSize->Full to resize each grid cell to its full width and height:

ImagePadding  (1)

By default, images are not padded:

Specify the same padding size on all sides:

Specify different padding in the horizontal and vertical directions:

Specify different padding for each side:

Padding  (1)

By default, with Padding->Automatic, images are padded with the same color as Background:

Use a different Background:

Specify a different padding:

Resampling  (1)

By default, a suitable resampling method is chosen:

Specify the resampling method:

Applications  (5)

Create a collage where images are weighted by their median intensity:

Create a collage of thumbnails of all test images in ExampleData:

Create a collage of a collection of Picasso's notable paintings:

Get images as well as their corresponding areas for these paintings:

Collage of paintings where each is shown at a size proportional to its original area:

Collage of images of sample buildings, weighted by their number of floors:

Collage of multiple barcodes:

Recognize all barcodes in the collage:

Properties & Relations  (3)

ImageCollage works with all kinds of images:

ImageAssemble requires regular images:

Use ConformImages to preprocess the set:

Use Grid to visualize images in a regular grid:

Use WordCloud to assemble words in 2D layout:

Possible Issues  (1)

If the image set exhibits some regularity, the weights may be altered to create a more packed collage:

Explicitly set all weights to 1 to avoid area differences:

Neat Examples  (3)

Create a country population visualization by showing each country flag, weighted by its population:

Select countries that contribute to at least 0.2 percent of the world population:

Create the collage:

Make a collage of 20 disks of random color and size:

Nestedly apply ImageCollage:

Wolfram Research (2014), ImageCollage, Wolfram Language function, (updated 2015).


Wolfram Research (2014), ImageCollage, Wolfram Language function, (updated 2015).


@misc{reference.wolfram_2020_imagecollage, author="Wolfram Research", title="{ImageCollage}", year="2015", howpublished="\url{}", note=[Accessed: 21-January-2021 ]}


@online{reference.wolfram_2020_imagecollage, organization={Wolfram Research}, title={ImageCollage}, year={2015}, url={}, note=[Accessed: 21-January-2021 ]}


Wolfram Language. 2014. "ImageCollage." Wolfram Language & System Documentation Center. Wolfram Research. Last Modified 2015.


Wolfram Language. (2014). ImageCollage. Wolfram Language & System Documentation Center. Retrieved from