Total Variation Denoising
Rudin, Osher and Fatemi proposed it in 1992: noise is removed by minimizing the total amount of intensity change, which flattens regions while keeping edges. Pushed hard, it yields a posterized, cartoon-like appearance.
Spot it
- 01Flat piecewise regions
- 02Sharp clean edges
- 03Staircase-like gradients
Why it happens
The algorithm finds the image closest to the noisy input that has the smallest sum of gradient magnitudes, favoring flat regions separated by clean jumps.
Recreate it in Pixio
Image edit prompt“Apply strong total variation denoising to this image: noise and texture removed, the picture reduced to flat, piecewise-constant regions with crisp clean edges, smooth gradients turning into subtle staircase steps, a slightly cartoonish look. Keep the original subject and composition.”
Open PixioUpload your image and pick an image-edit model.