Photo Humanizer forAI Images.
Run the 10-layer pipeline locally in your browser. Seven visible passes handle the core detector signals, while three reserved refinement layers stay internal - output reads as a real photo.
Photo Humanizer is sign-in only
The 10-layer humanization pipeline runs entirely in your browser — but you need a free account to start a session and track your weekly credits.
3 free credits every Monday. No card.
100% browser-only · your files never leave your deviceFree account · no credit card required
AI original vs humanized output.
Same pixels to your eye. A completely different statistical fingerprint to every detector.
Pipeline pillars
What runs under the hood
PRNU Sensor Fingerprint
A synthesized Photo Response Non-Uniformity pattern is injected so detectors classify your image as captured, not generated.
FFT Frequency Disruption
Diffusion models leave periodic artefacts in the frequency domain. Our FFT pass surgically scrambles those bands while staying perceptually invisible.
Real-Camera EXIF
iPhone, Pixel, Samsung and a generic DSLR profile — each with physically consistent ISO, aperture, focal length and GPS jitter.
Provenance Metadata Removal
Embedded provenance metadata — including Adobe C2PA manifests — is dropped on re-encode. An experimental SynthID-disruption pass is available but not guaranteed.
Adaptive 10-Layer Stack
Texture perturbation, colour decorrelation, chroma 4:2:0, JPEG double-compression, film grain, blur+sharpen — calibrated per image.
Zero-Upload Privacy
The full pipeline runs in a Web Worker on your device. Images never touch our servers.
How it runs
Three steps, all local
Drop image
Drag any AI-generated PNG, JPEG or WebP onto the canvas — single file or batch.
Local processing
10 layers run inside your browser via Web Worker. Typical: 3–6 seconds per image.
Authentic export
Download a real-camera-equivalent JPEG with embedded EXIF, ready for any platform.
Detector signals
Before vs. after — detector signals
-
Pixel-noise correlationSuspiciousCamera-consistent
-
FFT high-frequency energyPeriodic peaksNatural roll-off
-
EXIF reliability scoreMissing / syntheticPhysically valid
-
PRNU residualNoneProfiled sensor
-
JPEG quantisationSingle-passDouble-compressed
-
Provenance metadata (C2PA)PresentRemoved on re-encode
FAQ
Common questions
Will humanizing hurt visual quality?
No. Every layer operates at perceptually invisible amplitudes (sub-2 LSB on 8-bit channels). Output typically stays within 1–2 SSIM points of the original.
Does my image get uploaded?
Never. The entire pipeline runs locally inside a Web Worker. You can disable your network and processing still works.
Which detectors does this target?
It's designed to reduce the signals used by tools such as Hive, Sightengine, Optic, Illuminarty and AI-or-Not. No tool guarantees 100% — detectors evolve weekly, and so do we.
What file formats are supported?
PNG, JPEG, WebP and HEIC inputs. Output is always a real-camera-equivalent JPEG with embedded EXIF.