Low-key interior lighting
Relight the room in a restrained low-key style with one narrow soft beam entering the existing left window, illuminating the sofa edge and coffee table while the rest falls into deep readable shadow.
Relight this exact burger with a single hard warm-white key light high at camera left, producing a crisp deliberate cast shadow to the right and strong but readable side-to-side contrast.
A food photo with readable volume and a surface for cast shadow.
Relight this exact burger with a single hard warm-white key light high at camera left, producing a crisp deliberate cast shadow to the right and strong but readable side-to-side contrast.
Synthetic hard-light treatment; shadows and reflections are inferred, not a measured studio setup.
A hard key light makes surface relief and cast shadows prominent. Choose a direction that supports the subject's shape instead of darkening whichever side is easiest to regenerate.
Check the shadow's origin and the brightness of opposing surfaces. If important food or product details disappear, soften contrast locally. The edit does not reconstruct the original studio setup.
/LOWKEY — A low-key image emphasizes a dark tonal range, which need not use the same hard-light shadow structure.
These are editing suggestions, not additional test observations. Read the practical guide or use the before/after checklist.
Relight this exact burger with a single hard warm-white key light high at camera left, producing a crisp deliberate cast shadow to the right and strong but readable side-to-side contrast. Synthetic hard-light treatment; shadows and reflections are inferred, not a measured studio setup.
Input advice and use cases are editorial guidance, not additional test results.
Hard left light and crisp right cast shadow clearly demonstrate relighting, with original burger and plate recognizable. Minor seed and filling detail variation.
2026-09-24 · Work built-in image_gen
These notes describe the named runs, which may use a different subject from the comparison above. They do not guarantee results on another input or a future model. See our testing method.