Photo Colorizer

A black-and-white photograph records light, not color, so any color added afterwards is an inference. That is worth saying plainly before anything else. What the Photo Colorizer does is make that inference carefully: varied and believable skin tones, natural hair and eye colors, fabric, sky, foliage and brick that look the way those materials tend to look, with no color spilling across an edge into the next thing. Underneath, the luminance, the contrast and the grain of the original are kept, so the colorized version still reads as the same photograph. It is a way of seeing a familiar print differently, not a record of what was there. The original remains what it was, and the colorized version sits beside it as an interpretation.

The Era palette select shapes the inference. Match the era lets the AI read the clothing, cars and print style and pick a period-plausible palette on its own. The named decades push it deliberately: the 1900s to 1920s option gives the muted, slightly desaturated look of hand-tinted prints, the 1930s and 1940s a restrained early-color palette, the 1950s the rich warmth people associate with Kodachrome prints, the 1960s something brighter and cooler, the 1970s warm and earthy, the 1980s vivid and contrasty. The Known colors box is the most useful field on the page. If you remember that her dress was pale blue or the car was dark green, say so, and the AI treats it as fact rather than guesswork. Family memory is the best color reference there is.

Sepia and brown-toned prints colorize as readily as neutral ones, since the tone is treated as a tint to be replaced rather than as image color. Upload the largest scan you have. Results arrive in seconds and follow the shape of your upload, and if the print is also scratched or faded it is usually better to restore it first and colorize the clean version, which the Studio lets you do in one chain. Groups colorize well, since each person is given their own skin tone rather than a shared one. Interiors with many small objects are where the AI has to guess most, so a note about the wallpaper or the sofa goes a long way in a living-room photograph from the 1950s.