Files

T

Larry Gritz 1e8b0de711 admin: Update some license notices (#4955 )

Result of our annual license scan.

* Add license notices some places we had missed (mostly documentation).
* Double check some of the code files that are still marked as a mix of
Apache 2.0 + BSD, see if there are any remaining lines that are from
authors that hadn't relicensed their contributions -- there were a few
that had none of those old lines left, so they can be changed to
advertise that they are purely Apache 2.0.

Signed-off-by: Larry Gritz <lg@larrygritz.com>

2025-11-17 10:24:43 -08:00

2.4 KiB

Raw Permalink Blame History

OpenImageIO Usage Quick Start

Note

For information about how to install or build OpenImageIO, please see the INSTALL guide. This "Usage Quick Start" guide is intended to help you get started using OpenImageIO once you have it installed.

To illustrate how to do a simple common image operation -- reading a TIFF file and writing it back as an OpenEXR file, with all necessary conversions -- we will show how this can be accomplished through several different OpenImageIO APIs and programs.

In all cases, we will assume we have a file called input.tif and we wish to convert it to output.exr.

Using `oiiotool` command-line utility

$ oiiotool input.tif -o output.exr

High-level ImageBuf API

In C++:

#include <OpenImageIO/imagebuf.h>

void main()
{
    OIIO::ImageBuf image("input.tif");
    image.write("output.exr");
}

In Python:

import OpenImageIO as oiio

image = oiio.ImageBuf("input.tif")
image.write("output.exr")

Low level ImageInput/ImageOutput API

In C++:

#include <OpenImageIO/imageio.h>
#include <cstddef>
#include <memory>

void main()
{
    auto input = oiio.ImageInput.open("input.tif")
    OIIO::ImageSpec spec = input.spec()
    size_t nbytes = spec.image_bytes();
    std::unique_ptr<std::byte[]> pixels = new std::byte[nbytes];
    input->read_image(0, 0, 0, spec.nchanels, OIIO::TypeUnknown, pixels.get());
    input->close();

    auto output = oiio.ImageOutput.create("output.exr");
    output->open("output.exr", spec);
    output->write_image(TypeUnknown, pixels.get());
    output->close();
}

In Python:

import OpenImageIO as oiio

input = oiio.ImageInput.open("input.tif")
spec = input.spec()
pixels = input.read_image()
# Note: read_image will return a NumPy multi-dimensional array holding
# all the pixel values of the image.

output = oiio.ImageOutput.create("output.exr")
output.open("output.exr", spec)
output.write_image(pixels)
# Note: write_image expects a NumPy multi-dimensional array holding
# all the pixel values of the image.

Next Steps

This is just a simple example to give you a flavor of the different major interfaces. For more advanced usage, you may want to explore the documentation.

2.4 KiB Raw Permalink Blame History