VROF

AntiFake

Detects forged and altered face videos in real time. Developed under the Ministry of SMEs and Startups (중소벤처기업부) Startup Growth Technology Development Project (창업성장기술개발사업), with eight items measured by an external testing institution.

중소벤처기업부 창업성장기술개발(R&D) · Duration 2023.05 – 2024.07

Project overview

Development period
15 months
Project status
Completed

A technology that detects forged and altered face videos in real time with AI,
developed to completion as a national R&D project.

Non-face-to-face account opening, remote identity verification, video interviews. As more procedures identify a person by their face, procedures that confirm the face itself is real have become just as necessary.

Technology that analyzes forgery and alteration after the fact already exists. What this project targets is real time. If the verdict does not come at the very moment the verification procedure is in progress, there is nothing left to do with it.

Verified performance

Test items
8
Verdict
All 8 items PASSED
Test
2024.04

Numbers measured by an external institution.

The Gumi Electronics and Information Technology Research Institute (구미전자정보기술원, GERI) measured eight items, and all eight exceeded their targets.

Gumi Electronics and Information Technology Research Institute (구미전자정보기술원, GERI) performance evaluation results — 8 items, split by measurement environment
Evaluation item Unit Target Measured Verdict
Desktop measurements GPU RTX 3090 · CPU Intel Core i9-10900F · RAM 32GB
Face image feature point extraction accuracy % 95 or higher 99 PASS
AntiDeepFake face extraction accuracy % 93 or higher 99.84 PASS
Face image feature point extraction time ms 33.3 or lower 2.043 PASS
AntiDeepFake face extraction time ms 222 or lower 7.283 PASS
Smartphone measurements Samsung Galaxy S21 Ultra (Android) · iPhone 13 Pro Max (iOS) · 5G
Face image feature point extraction speed on mobile FPS 20 or higher 70.3 Android 74.3 iOS PASS
AntiDeepFake face extraction speed on mobile FPS 7 or higher 59.2 Android 41.8 iOS PASS
Processor usage during feature point extraction on mobile % 60 or lower 46.8 Android 40.8 iOS PASS
Processor usage during AntiDeepFake detection on mobile % 90 or lower 48.2 Android 66.68 iOS PASS
Testing institution
구미전자정보기술원(GERI)
Test period
2024.03.29 – 2024.04.18
Issue date
2024.04.26
Test location
주식회사 브이로프
Dataset
FaceForensics++ (DeepFakes · Face2Face · FaceSwap · NeuralTextures)
Measurement method
Average of 30 samples × 5 repeated measurements
Cover of the performance evaluation certificate issued by the Gumi Electronics and Information Technology Research Institute (구미전자정보기술원, GERI). It lists the requesting company VROF Inc. (주식회사 브이로프), the test item: AI-applied real-time detection technology for forged and altered face videos (AntiFake), and the number of test items: 8.
Cover of the performance evaluation certificate.

Application points

Scenario
4

Where it applies.

  • eKYC · Non-face-to-face identity verification

    Determines whether the face video submitted during account opening and identity verification is from an actual recording.

  • Remote identity verification

    Checks whether the other party is being captured in real time during an identity verification conducted over video.

  • Video evidence verification

    Applied to procedures that must determine whether submitted video material has been forged or altered.

  • Content platform filtering

    Applied to services that must filter out forged or altered videos at the upload stage.

AntiFake integration structure Video input passes through AntiFake's real-time detection. If judged real, it moves on to the existing identity verification procedure and is approved; if judged fake, it is blocked before entering the procedure. Video input Camera · upload AntiFake Real-time forgery check Frame by frame Existing ID check Face match · ID Approve Block · retry request Before entry Real Fake Only this block is added

Product

Released
2024.02.06
Platform
iOS

Available for free on the App Store.

Released2024.02.06
Supported languagesKorean · English · Traditional Chinese · Japanese
AvailabilityDownloaded in 25 countries (App Store, as of 2024.09.25)
Download App Store
A collection of AntiFake app screens in English, Traditional Chinese, and Japanese. For each language, the password recovery, sign-up, login, and start screens are laid out side by side.
Screens in English, Traditional Chinese, and Japanese.

Embedded use in corporate systems, technology transfer, and licensing are negotiated separately. Tell us about the environment you want to apply it to, and we will first lay out the feasible options.

Development history

Support program
Startup Growth Technology Development (창업성장기술개발)
Government support
KRW 120 million

Carried out as a national R&D project.

Project titleAI 적용, 얼굴 위변조 영상 실시간 탐지기술 개발
Support program중소벤처기업부 창업성장기술개발(R&D)
Specialized agency중소기업기술정보진흥원(TIPA)
Project period2023.05 – 2024.07
Principal investigator (CEO)
Total R&D cost KRW 150 million (Government support KRW 120 million)
Development deliverablesAntiFake (Software)
Certificate of project performance (연구과제수행확인서). Program: Startup Growth Technology Development Project (창업성장기술개발사업, Didimdol track), total research start date 2023-05-01, total research cost KRW 150,000,000, issued by the Korea Technology and Information Promotion Agency for SMEs (중소기업기술정보진흥원, TIPA).
Certificate of project performance — Startup Growth Technology Development (Didimdol track). Total research cost confirmed as KRW 150 million.
Certificate of project performance (과제수행 확인증). Program: Small and Medium Enterprise Technology Innovation Development Project (중소기업기술혁신개발사업, market-responsive track), issued by the head of the Korea Technology and Information Promotion Agency for SMEs (중소기업기술정보진흥원, TIPA).
Certificate of project performance — Small and Medium Enterprise Technology Innovation Development Project. A project the CEO carried out as project lead at Eenter Co., Ltd. ((주)이엔터).
Technology data escrow certificate (기술자료 임치증). Contract no. 2024-02-32-5003, escrowed material title 「AI 적용, 얼굴 위·변조 영상 실시간 탐지기술 개발」, escrow period 2024.09.20 – 2025.09.19, developer VROF Inc. (주식회사 브이로프), issuing organization: Korea Foundation for Cooperation of Large·Small Business, Agriculture and Fishery (대·중소기업·농어업협력재단).
Technology data escrow certificate (대·중소기업·농어업협력재단). Core source code was escrowed so that the technology copyright is protected.

This technology is the result of the Startup Growth Technology Development Project (창업성장기술개발사업, Didimdol track) conducted by the Ministry of SMEs and Startups (중소벤처기업부).

Patents

Patents
2 registered · 1 pending
Inventor
Named on all

We state the rights holder and the inventor together.

The CEO is listed as an inventor on all three.

Type Title · number Filed Registered Rights holder · inventor
Registered System and Method for Producing Special Effects through Face Recognition 10-2529209 2021.09.16 2023.04.28 주식회사 이엔터 Inventor and others
Registered Distributed Processing System for Real-time Deepfake Image Detection in Mobile Environments 10-2896820 2022.12.28 2025.12.02 주식회사 이엔터 Inventor and others
Filed AfterRemaker (Class 09) 10-2024-0022442 2024.02.16 Under examination 주식회사 브이로프 Inventor and others
Patent registration certificate no. 10-2529209. Invention title: 「System and Method for Producing Special Effects through Face Recognition」, patent holder: Eenter Co., Ltd. (주식회사 이엔터).
Patent no. 10-2529209 — special effect production system using face recognition
Patent registration certificate no. 10-2896820. Invention title: 「Distributed Processing System for Real-time Deepfake Image Detection in Mobile Environments」, patent holder: Eenter Co., Ltd. (주식회사 이엔터).
Patent no. 10-2896820 — distributed processing for real-time deepfake detection on mobile
Patent application number notice. Application no. 10-2024-0022442, filing date 2024.02.16, applicant VROF Inc. (주식회사 브이로프), invention title: 「Artificial Intelligence-based Device and Method for Detecting Forged Images」.
Application 10-2024-0022442 — filed by VROF, under examination

Long-running collection

Verification
26 tests
Criterion
Measured against the actual code

Collectors are built on the assumption that they will stop.

A collection run that spans days is bound to break somewhere in the middle. Responses stop coming, formats break, computers shut down. The stopping itself cannot be prevented. Instead, from the start we build in a mechanism that resumes from where it stopped and fills in only what was missed — so a run that stopped midway is never handed over half-done; we fill the gaps and deliver the rest.

  1. 01

    We read the values the page already carries, not the screen

    When you point at the screen structure with selectors, a site redesign can quietly bring in different values.

  2. 02

    It remembers where it stopped

    Large-scale collection is bound to break midway. When it restarts, it resumes from the saved progress point.

  3. 03

    Re-fetching never overwrites what was already collected

    It merges by unique identifier. Data already collected stays intact, and only duplicates are filtered out.

  4. 04

    Failed items are kept in a list, and only those are re-fetched

    Addresses that failed collection are recorded separately, and a script re-fetches only that list.

  5. 05

    When it gets stuck, it switches paths and continues

    When no response comes or a response arrives in a broken format, it switches to another path and continues from that point.

  6. 06

    Handed over in a form that runs without a development environment

    It is built as an executable so the person in charge can use it without installing Python.

The longest-running collection system we operated covered 40 or more sites for 5 years. It was not built, handed over, and done — this runtime comes from continually fixing it as target sites changed or broke.

Technologies

Technologies
11
CEO Experience
23 years in development

Technologies worked with across 23 years of development

The CEO directly designs and implements. Depending on project scale, developers are directly selected to join, and requirements are never subcontracted wholesale to third parties.

  • Python
  • PHP
  • Linux
  • HTML
  • Docker
  • AWS
  • EC2
  • GitHub
  • Redis
  • MongoDB
  • Node.js

A permanent team of 3 works directly, with no subcontracting to third parties.

Contact

Reply
Within 1 business day

If you are considering adoption or technical collaboration.

Tell us the target environment and what needs to be detected, and we will first lay out which approaches are feasible.

We offer consultations to help you save time Free consultation · Replies within 1 business day

Contact