OpenEXR

OpenEXR provides the specification and reference implementation of the EXR file format, the professional-grade image storage format of the motion picture industry.

The purpose of EXR format is to accurately and efficiently represent high-dynamic-range scene-linear image data and associated metadata, with strong support for multi-part, multi-channel use cases.

OpenEXR is widely used in host application software where accuracy is critical, such as photorealistic rendering, texture access, image compositing, deep compositing, and DI.

OpenEXR Project Mission

The goal of the OpenEXR project is to keep the EXR format reliable and modern and to maintain its place as the preferred image format for entertainment content creation.

Major revisions are infrequent, and new features will be carefully weighed against increased complexity. The principal priorities of the project are:

• Robustness, reliability, security
• Backwards compatibility, data longevity
• Performance - read/write/compression/decompression time
• Simplicity, ease of use, maintainability
• Wide adoption, multi-platform support - Linux, Windows, macOS, and others

OpenEXR is intended solely for 2D data. It is not appropriate for storage of volumetric data, cached or lit 3D scenes, or more complex 3D data such as light fields.

The goals of the Imath project are simplicity, ease of use, correctness and verifiability, and breadth of adoption. Imath is not intended to be a comprehensive linear algebra or numerical analysis package.

Python Module

The OpenEXR python module provides full support for reading and writing all types of .exr image files, including scanline, tiled, deep, mult-part, multi-view, and multi-resolution images with pixel types of unsigned 32-bit integers and 16- and 32-bit floats. It provides access to pixel data through numpy arrays, as either one array per channel or with R, G, B, and A interleaved into a single array RGBA array.

Project Governance

OpenEXR is a project of the Academy Software Foundation. See the project's governance policies, contribution guidelines, and code of conduct for more information.

Quick Start

The "Hello, World" image writer:

# Generate a 3D NumPy array for RGB channels with random values
height, width = (20, 10)
RGB = np.random.rand(height, width, 3).astype('f')

channels = { "RGB" : RGB }
header = { "compression" : OpenEXR.ZIP_COMPRESSION,
           "type" : OpenEXR.scanlineimage }

with OpenEXR.File(header, channels) as outfile:
    outfile.write("readme.exr")

Or alternatively, construct the same output file via separate pixel arrays for each channel:

# Generate arrays for R, G, and B channels with random values
height, width = (20, 10)
R = np.random.rand(height, width).astype('f')
G = np.random.rand(height, width).astype('f')
B = np.random.rand(height, width).astype('f')
channels = { "R" : R, "G" : G, "B" : B }
header = { "compression" : OpenEXR.ZIP_COMPRESSION,
           "type" : OpenEXR.scanlineimage }

with OpenEXR.File(header, channels) as outfile:
    outfile.write("readme.exr")

The corresponding example of reading an image is:

with OpenEXR.File("readme.exr") as infile:

    RGB = infile.channels()["RGB"].pixels
    height, width, _ = RGB.shape
    for y in range(height):
        for x in range(width):
            pixel = tuple(RGB[y, x])
            print(f"pixel[{y}][{x}]={pixel}")

Or alternatively, read the data as separate arrays for each channel:

with OpenEXR.File("readme.exr", separate_channels=True) as infile:

    header = infile.header()
    print(f"type={header['type']}")
    print(f"compression={header['compression']}")

    R = infile.channels()["R"].pixels
    G = infile.channels()["G"].pixels
    B = infile.channels()["B"].pixels
    height, width = R.shape
    for y in range(height):
        for x in range(width):
            pixel = (R[y, x], G[y, x], B[y, x])
            print(f"pixel[{y}][{x}]={pixel}")

To modify the header metadata in a file:

with OpenEXR.File("readme.exr") as f:
    
    f.header()["displayWindow"] = ((3,4),(5,6))
    f.header()["screenWindowCenter"] = np.array([1.0,2.0],'float32')
    f.header()["comments"] = "test image"
    f.header()["longitude"] = -122.5
    f.write("readme_modified.exr")

    with OpenEXR.File("readme_modified.exr") as o:
        dw = o.header()["displayWindow"]
        assert (tuple(dw[0]), tuple(dw[1])) == ((3,4),(5,6))
        swc = o.header()["screenWindowCenter"]
        assert tuple(swc) == (1.0, 2.0)
        assert o.header()["comments"] == "test image"
        assert o.header()["longitude"] == -122.5

Note that OpenEXR's Imath-based vector and matrix attribute values appear in the header dictionary as 2-element, 3-element, 3x3, 4x4 numpy arrays, and bounding boxes appear as tuples of 2-element arrays, or tuples for convenience.

To read and write a multi-part file, use a list of Part objects:

height, width = (20, 10)
Z0 = np.zeros((height, width), dtype='f')
Z1 = np.ones((height, width), dtype='f')

P0 = OpenEXR.Part({}, {"Z" : Z0 })
P1 = OpenEXR.Part({}, {"Z" : Z1 })

f = OpenEXR.File([P0, P1])
f.write("readme_2part.exr")

with OpenEXR.File("readme_2part.exr") as o:
    assert o.parts[0].name() == "Part0"
    assert o.parts[0].width() == 10
    assert o.parts[0].height() == 20
    assert o.parts[1].name() == "Part1"
    assert o.parts[1].width() == 10
    assert o.parts[1].height() == 20

Deep data is stored in a numpy array whose entries are numpy arrays. Construct a numpy array with a dtype of object, and assign each entry a numpy array holding the samples. Each pixel can have a different number of samples, including None for no data, but all channels in a given part must have the same number of samples.

height, width = (20, 10)

Z = np.empty((height, width), dtype=object)
for y in range(height):
    for x in range(width):
        Z[y, x] = np.array([y*width+x], dtype='uint32')

channels = { "Z" : Z }
header = { "compression" : OpenEXR.ZIPS_COMPRESSION,
           "type" : OpenEXR.deepscanline }
with OpenEXR.File(header, channels) as outfile:
    outfile.write("readme_test_tiled_deep.exr")

To read a deep file:

with OpenEXR.File("readme_test_tiled_deep.exr") as infile:

    Z = infile.channels()["Z"].pixels
    height, width = Z.shape
    for y in range(height):
        for x in range(width):
            for z in Z[y,x]:
                print(f"deep sample at {y},{x}: {z}")

Community

• Ask a question:

  • Email: openexr-dev@lists.aswf.io

  • Slack: academysoftwarefdn#openexr

• Attend a meeting:

  • Technical Steering Committee meetings are open to the public, fortnightly on Thursdays, 1:30pm Pacific Time.

  • Calendar: https://lists.aswf.io/g/openexr-dev/calendar

  • Meeting notes: https://wiki.aswf.io/display/OEXR/TSC+Meetings

• Report a bug:

  • Submit an Issue: https://github.com/AcademySoftwareFoundation/openexr/issues

• Report a security vulnerability:

  • Email to security@openexr.com

• Contribute a Fix, Feature, or Improvement:

  • Read the Contribution Guidelines and Code of Conduct

  • Sign the Contributor License Agreement

  • Submit a Pull Request: https://github.com/AcademySoftwareFoundation/openexr/pulls

Resources

• Website: http://www.openexr.com
• Technical documentation: https://openexr.readthedocs.io
• Porting help: OpenEXR/Imath Version 2.x to 3.x Porting Guide
• Reference images: https://github.com/AcademySoftwareFoundation/openexr-images
• Security policy: SECURITY.md
• Release notes: CHANGES.md
• Contributors: CONTRIBUTORS.md

License

OpenEXR is licensed under the BSD-3-Clause license.