You have an old, damaged black-and-white photo. Should you restore the damage first, then colorize? Or colorize first, then restore? The order changes the result. Here's the correct pipeline.
You have a 70-year-old black-and-white photo. It is faded, scratched, and has a crease running through the middle. You want to: restore the damage (remove scratches, fix the crease, enhance contrast) and colorize the photo (add realistic color). You have a photo restorer and a colorizer. Which one do you use first?
The answer: restore first, then colorize. The order matters more than you think — and using the wrong order produces worse results that cannot be fixed without starting over. Here is the correct pipeline and why the order matters.
Reason 1: Damage confuses the colorizer. A scratch across a person's face is a high-contrast artifact. The AI colorizer sees the scratch as a feature of the image and assigns it a color — usually a skin tone or a gray tone. The scratch becomes a colored line across the face. The colorization locks in the scratch as a permanent feature. If you colorize first and then try to remove the scratch, the repair must match the surrounding colored area — which is harder than repairing a grayscale area. Restoring first removes the scratch before it can be colorized.
Reason 2: Fading affects color accuracy. A faded photo has low contrast. The AI colorizer uses contrast information to determine color boundaries — the edge between a person's hair and the background, the boundary between a shirt and a jacket. Low contrast makes these boundaries ambiguous. The colorizer guesses where the boundaries are and applies colors based on those guesses. The result: colors bleeding across boundaries. Restoring first enhances the contrast, which gives the colorizer cleaner boundaries to work with.
Reason 3: Stains and discoloration mislead the colorizer. A yellowed or sepia-toned photo has a color cast. The AI colorizer sees the yellow or sepia as the actual color of the image and may incorporate it into the colorization. A yellow-stained white shirt becomes a yellow shirt. A sepia-toned sky becomes a brown sky. Restoring first neutralizes the color cast, giving the colorizer a clean monochrome image to work from.
Step 1: Scan the original photo at the highest possible resolution. 600 DPI minimum. The original scan is the master file. All restoration and colorization are derived from this master. Never modify the master directly.
Step 2: Restore with the photo restorer. Use the photo restorer to: remove scratches, dust, and physical damage, fix creases and folds, enhance contrast and sharpness, and neutralize color casts (sepia, yellowing). The output is a clean, high-contrast black-and-white image.
Step 3: Colorize with the colorizer. Feed the restored black-and-white image to the colorizer. The clean image gives the AI the best possible input. The colors will be more accurate, the boundaries will be cleaner, and the result will look more natural.
Step 4: Review the colorized result. Check for: color bleeding across boundaries, colors that look unnatural (a green face, a purple sky), and artifacts introduced by either the restoration or the colorization. The colorized result is a derived work. The original scan is the master. Keep both.
There is one exception: if the photo has severe fading where you cannot tell what the image depicts, colorizing first might help you identify the content — which parts are faces, which are clothing, which are background. The colorization adds information that helps you see the image more clearly. Then you can restore the original black-and-white scan more effectively with that understanding. Then you can colorize the restored version for the final output. In this case, the pipeline becomes: colorize (for analysis) → restore (with the understanding gained) → colorize (for final output). But this is the exception, not the rule.
For 95% of old photos, the pipeline is: restore first, then colorize. The order is the difference between a natural-looking colorized photo and one where the scratches and stains are now permanently colored. Restore at photo restorer, then colorize at colorizer. The right order. The right result.
B&W Photo Colorizer
Bring black and white photos to life with natural, vibrant AI colorization.
Photo Restorer
Restore and colorize old, blurry, or damaged photos.
Image Upscaler
Increase image resolution up to 4x with Real-ESRGAN AI upscaling. Dedicated Photo and Anime modes for different image types. Choose 2x or 4x upscaling factor. Enhances old photos, AI-generated images, and low-res pictures to HD quality without losing detail. Perfect for printing and digital displays.