Watermark segmentation

Do any of these watermark removal systems support simple &quot;training&quot; on multiple images with identical watermarks? Having multiple example images with consistent watermarks should make removing watermarks much easier than trying to remove one with no context.<p>I haven&#x27;t found a tool that implements the techniques described in this Google paper from 8 years ago: <a href="https:&#x2F;&#x2F;watermark-cvpr17.github.io&#x2F;" rel="nofollow">https:&#x2F;&#x2F;watermark-cvpr17.github.io&#x2F;</a>

Honest question, is there even a <i>legitimate</i> use for this specific tech?

Their first example at <a href="https:&#x2F;&#x2F;www.clear.photo&#x2F;en" rel="nofollow">https:&#x2F;&#x2F;www.clear.photo&#x2F;en</a> is absolutely terrible. I assume a showcase would show &quot;good&quot; results, but they display a complete failure.<p>- Incorrectly identifies areas for inpainting. You can see this with the figure, a lot of detail, not obscured by the watermark, is erased and then redrawn. This leads to a totally distorted look. The belt just disappears into nothing, the cloth just becomes a gradient, where a crisp line used to be.<p>- Low quality inpainting. Even the inpainting is done terribly. This looks like something done with some very simple diffusion based inpainting. Absolutely not state of the art.

And some people call generative AI nothing but a copyright laundry…

This is technicaly impressive, but I wonder if this could be put to a use which is generally more constructive. Like maybe removing stains from scans or red eye from pictures.

I&#x27;m surprised there isn&#x27;t a readily available water-mark remover at this point. A synthetic training set for such a model could be created trivially.