Software:LUT Generator
LUT Generator is a web-based tool for developing a colour treatment from a source image and a visual reference. Its workflow supports photographs and frames from standard-dynamic-range video, interactive colour adjustment, and export of a reusable .cube lookup-table file. The image-processing workflow runs locally in the browser. An optional AI colour assistant provides another way to explore a creative direction.
Inputs and workflow
The input consists of source material and a reference image. The source represents the photograph or video frame to be treated, while the reference communicates a desired relationship between colours, contrast, highlights, and shadows. A reference is a guide rather than a guarantee that two scenes with different lighting or subjects will become visually identical.
A typical workflow begins by choosing a representative source frame and a reference with a useful tonal range. The user examines the proposed colour relationship and adjusts the result before exporting a lookup table. A neutral object or a face can provide a practical comparison point, although a reference does not establish an objectively correct appearance for every subject.
Lookup-table export
A colour lookup table describes a mapping between input and output colour values. A .cube file can carry that mapping into editing applications that support the format. Compatibility with the file format does not, by itself, establish that a treatment is appropriate for a particular colour-management pipeline. The technical interpretation of source footage and the position of the creative adjustment remain relevant.
A look developed from display-ready photographs or standard-dynamic-range frames is distinct from a camera-specific conversion of log footage. Exposure and white-balance corrections may therefore be needed before the creative treatment is applied. The exported file provides a repeatable starting point rather than a replacement for individual shot correction.
Evaluation
A reusable treatment can be evaluated with several images rather than only the frame used to develop it. Bright exteriors, dark interiors, faces, neutral surfaces, and saturated objects expose different behaviour. Comparisons can reveal clipped highlights, compressed shadows, unexpected changes to skin colour, or excessive saturation. These observations help distinguish a problem affecting the shared treatment from a correction needed by one unusual shot.
For video, the edited sequence provides another level of evaluation. Adjacent shots may reveal changes in colour balance that are less obvious when still frames are examined separately. Consistent source preparation and a documented processing order make the result easier to interpret and revise.
Scope and limitations
Reference-based colour matching concerns the visual treatment of existing material. It does not recreate the lighting, subject, composition, or dynamic range of the reference. A lookup table also cannot recover detail that was not captured in the original source. Users still need to inspect the result in the intended application and viewing context.
The optional AI assistant is a creative aid rather than an independent validation of a grade. Reference images also carry their own permissions and privacy considerations. Choosing relevant material and reviewing the output remain parts of the workflow, irrespective of whether automated assistance is used.
External link
- lut generator from image — project website
