allenk/GeminiWatermarkTool

Offline Gemini image watermark restoration with reverse alpha blending. Native C++ GUI/CLI, batch processing, and optional GPU denoise.

View on GitHub ↗Jump to charts ↓Open shareable report →

allenk/GeminiWatermarkTool has 3k+ stars and 250+ forks on GitHub, is written in C++, and was last pushed on Jul 28, 2026. Data as of . Signed-in members get hourly updates — create a free account.

Summary Information

Updated 2 hours ago
Added to GitGenius on September 22nd, 2026
Created on December 12th, 2025
Open Issues & Pull Requests: 10 (+0)
GitHub issues: Enabled
Number of forks: 271
Total Stargazers: 3,130 (+0)
Total Subscribers: 8 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 34.8 hours
Mean response time: 2.2 days
90th percentile: 4.6 days
Tracked items: 30

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 9
New in 7 days: 0
Closed in 7 days: 0
Avg open age: 127 days
Stale 30+ days: 9
Stale 90+ days: 6

Recent activity

Opened in 7 days: 0
Closed in 7 days: 0
Comments in 7 days: 0
Events in 7 days: 0

Top labels

No label distribution available yet.

Most active issues this week

Sign in to see which issues are moving.
Sign in

Detailed Description

Gemini Watermark Tool is a desktop application that removes Gemini AI watermarks from images using reverse alpha blending.

The tool solves the problem of unwanted watermarks on Gemini-generated images by mathematically inverting the alpha blending equation that creates the watermark, rather than relying on generative inpainting. It uses calibrated reverse-alpha masks for both standard and half-scale watermark variants to reconstruct the underlying image detail. The approach is deterministic and works offline without requiring cloud services or AI hallucination.

The tool suits developers and users who need reliable, reproducible watermark removal without external dependencies. It works as a portable executable with no runtime requirements, making it suitable for automation, batch processing, and pipeline integration. The project offers both a native C++ GUI for interactive use and a command-line interface for scripting. It supports automatic detection and removal of both legacy and Gemini 3.5 watermark profiles, with optional GPU-accelerated denoising for additional cleanup. The README does not compare this approach to alternatives.

Development shows active maintenance with recent updates addressing edge cases. The project added support for free-tier half-scale watermarks with improved detection robustness on complex backgrounds. Support for the Gemini 3.5 watermark profile was implemented with automatic fallback to legacy detection when needed. The maintainer has extended the reverse alpha blending technique to a companion video watermark removal tool, demonstrating continued evolution of the core method.