Cork/Face Presentation Attack Detection

RF tool based on CC1101 module and Arduino Pro Micro 8VMHz/3.3V. Allows using CLI with human readable commands to control CC1101 board over USB interface. Putty or any other serial terminal can be used. It has similar functionality to YardStick One but is cheaper and doesn't need specialized software. Allows for RF jamming and simple replay attack

Intercepting Bluetooth device communication and simulating packet responses of an iPhone from a Raspberry Pi 3

Example of using CC1101 module with Arduino Pro Micro ( ATMEGA32U4 ) for radio jamming

Very simple 433MHz (EU/ASIA) 315 MHz (US) RF jammer for keyless cars theft prevention

this is example of cloning my garage keys with Arduino Digispark and FS1000A transmitter. The keys sequences were recorded first using Universal Radio Hacker tool

Liveness Detection, Deepfake Detection and PhotoVerify Face Match for KYC

BWS Android sample code for app integration

Zalo AI Challenge 2022 Liveness Detection

In this project, we have implemented a network to detect real faces against fake ones.

BWS HTML5 Unified User Interface

AdvSV stands as the first dataset developed specifically for evaluating Speaker Verification (SV) systems against adversarial attacks. It aims to benchmark the robustness of ASV models in the face of such attacks and offers vital resources for researchers to explore the characteristics of adversarial and replay attacks in this domain.

Deep learning-based audio spoofing attack detection experiments for speaker verification.

BWS iOS sample code for app integration

Baseline Experiments for ReMASC dataset.

The source code in the paper "When the Differences in Frequency Domain are Compensated: Understanding and Defeating Modulated Replay Attacks on Automatic Speech Recognition". This paper is published in the ACM Conference on Computer and Communications Security (CCS) 2020.

A nonce manager PHP library useful for preventing CSRF and replay attacks.

A PHP Library for Cryptographically Secure Token Generation and Management

Test of libSodium's ChaCha20-Poly1305 with increases security by deriving a different key per message

This project presents an SVM model trained to distinguish between Real human voices and Replayed human voice.