8000 GitHub - JiePKU/FSCA: Official code for Paper: "Are Your Images Used for Training? Auditing Data Provenance in Real-world Text-to-image Diffusion Models"
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content
/ FSCA Public

Official code for Paper: "Are Your Images Used for Training? Auditing Data Provenance in Real-world Text-to-image Diffusion Models"

1 star 0 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

JiePKU/FSCA

BranchesTags

Repository files navigation

Auditing Data Provenance in Real-world Text-to-Image Diffusion Models for Privacy and Copyright Protection

Quick Start

Requirements

  • Python 3.8.16
  • pytorch 1.12.1+cu113
  • numpy 1.20.3
  • torchvision 0.13.1+cu113
  • scikit-learn 1.2.2
  • diffusers 0.17.1
  • clip
  • datetime
  • pathlib
  • PIL

Quick Start

Download Stable Diffusion v1.5 and LAION-mi dataset

Extract text and image features

sh sh_extract.sh

Train and evaluate FSCA auditing model

sh fsca_command/sh_train_0.sh

About

Official code for Paper: "Are Your Images Used for Training? Auditing Data Provenance in Real-world Text-to-image Diffusion Models"

Resources

Readme
Activity

Stars

1 star

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages
0