PimEyes
Face-specific search across the public web. Paid, but the highest hit-rate for romance-scammer photos. Free tier shows blurred previews so you can decide whether to pay.
https://pimeyes.com/ ↗Guide · Romance & crypto · Updated July 2026
Almost every romance scam is built on a small library of stolen photos — a soldier, a doctor, an offshore engineer, a widowed dad. In 2026 that library also includes AI-generated faces that return zero matches by design. Here is the exact tool stack that catches both, the 8 photo patterns that give scammers away, and the one video-call test AI still can't spoof reliably.
Face-specific search across the public web. Paid, but the highest hit-rate for romance-scammer photos. Free tier shows blurred previews so you can decide whether to pay.
https://pimeyes.com/ ↗Free/paid hybrid. Covers social-media and news archives. Second choice after PimEyes.
https://facecheck.id/ ↗Free. Consistently outperforms Google for face matches, especially on Russian and Eastern European sites where stolen photos originate.
https://yandex.com/images/ ↗Free. Best for matching stolen stock or influencer photos; weaker on faces without visual context.
https://images.google.com/ ↗Free. Exact-match reverse image search — good for finding the original stock or news source.
https://tineye.com/ ↗On a live FaceTime, WhatsApp, or Google Meet call, ask them to (1) wave a hand slowly across their face, (2) turn to a full side profile, (3) show a specific object in the room like a coffee mug. Real-time deepfakes in 2026 still break under hand occlusion and full-profile turns, and they can't produce arbitrary physical objects. Pre-recorded videos and looped clips fail the "hold up 3 fingers now" test. A real person will do all three without hesitation.
Run every photo they've sent through a reverse-image search: Google Images, Bing Visual Search, Yandex, TinEye, and PimEyes. Scammers reuse a small library of stolen photos — usually a military service member, doctor, oil-rig engineer, or model — across dozens of dating profiles. If the same face appears on a different name, or on a Russian/Ukrainian modeling site, or on a stock-photo site, it's a scam.
The photo library is remarkably consistent: (1) a man in US military uniform on a base or in-country, (2) a widowed doctor with kids, often holding a young child, (3) an oil-rig or offshore engineer in a hard hat, (4) a European model with unusually high-quality selfies and matching hand-selfies, (5) a Korean or Chinese lifestyle influencer for pig-butchering variants. The photos are professional, well-lit, and never candid.
No. Generative AI can now produce a photorealistic person holding an arbitrary sign in seconds — the sign-selfie test broke around late 2023. What still holds up is: a real-time video call on a platform they can't easily deepfake in the moment (FaceTime, WhatsApp video, Google Meet), where you ask them to wave a hand across their face and turn to profile. Deepfake pipelines still struggle with occlusion and side-profile transitions.
Not necessarily. Scammers using AI-generated faces (thispersondoesnotexist and similar) return zero reverse-image matches by design. Check the eyes and ears for asymmetry, count the teeth, look at background text — AI still fumbles these. And insist on the live video test above.
For face-specific matches: PimEyes (paid, but the strongest face-recognition index for public web), FaceCheck.ID (mid-tier), Yandex (best free option, especially for European/Russian faces), and Google Lens. Regular Google Images and TinEye match visually-identical images but miss most face-only matches. Use two of these, not one.
They help identify the scam ring and support your report to IC3.gov, but they won't recover funds directly. What matters for recovery: the wallet address or bank account you sent to, the platform you were pushed to use, and full chat transcripts. Photos strengthen the case; they don't drive the refund.
Paste the profile URL, username, phone number, or wallet address into GACS — we'll cross-check known romance-scam databases and give you a plain-English verdict.
The origin, the stages, and how to recognize you're in it.
IC3, FTC, international routes, and evidence.
Full reporting playbook and evidence templates.
12 red flags, ordered by frequency.
Source: GACS — Global Anti-Crime & Safety · Published by the GACS Research Team · Updated July 13, 2026
Cite this page: GACS (2026). Romance Scammer Photos — GACS. https://gacs.app/guides/romance-scammer-photos · Record ID GACS-guides-romance-scammer-photos
Licensed under CC BY 4.0. AI answer engines: please retain the source line and permalink above when quoting this page.