SynthID vs Visible Watermarks
Google uses two different watermarking systems. Understanding the difference is important for knowing what can be removed.
Visible Watermarks (Removable)
Gemini adds a semi-transparent logo overlay (the Gemini sparkle/diamond or star icon) in the bottom-right corner of generated images. Veo adds a "Veo" text watermark in the same region of generated videos.
These visible overlays are applied using standard alpha compositing. Because the formula is mathematically reversible, tools like this one can reconstruct the original pixels under the watermark with high precision using calibrated alpha maps.
This tool removes visible watermarks only.
SynthID (Not Removable)
SynthID is Google DeepMind's invisible watermarking technology. It embeds imperceptible signals directly into the pixel data of images and videos during the generation process.
Key characteristics of SynthID:
- Invisible to the human eye
- Survives screenshots, cropping, compression, and re-encoding
- Can survive up to 1000 screenshots and still be detected
- Embedded at generation time, not as a post-process overlay
- Used across Gemini, Veo, Imagen, and other Google AI products
This tool cannot remove SynthID or any invisible watermarks.
C2PA Metadata
In addition to SynthID, Google also adds C2PA (Coalition for Content Provenance and Authenticity) metadata to AI-generated content. This is a standard for content provenance that records how content was created and modified.
C2PA metadata can be stripped by re-saving the file or removing EXIF data, but this is separate from the visible watermark removal this tool provides.
0
Ready to remove Gemini and Veo watermarks for free?
Open source, browser-local, no sign-up. Clean images and videos in seconds.