Generative-Content-Replacement

This is the interface for the trying generative content replacement (GCR), for CHI 2024 and SOUPS 2024 poster. For details of GCR implementation, please refers to our CHI 2024 paper "Examining Human Perception of Generative Content Replacement in Image Privacy Protection".

Interface Layout

Getting started

1. Prerequisite

A GPU with >24Gb VRAM, 32Gb DRAM, 100Gb Storage, Nodejs, Anaconda

2. Installation

git clone https://github.com/AnranXu/Generative-Content-Replacement.git

cd Generative-Content-Replacement
npm install
conda create -n GCR python=3.10
conda activate GCR
pip install -r requirements.txt
conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia
cd backend
mkdir pretrained_models
cd pretrained_models
wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth

3. Setting A Development Server To Try GCR

Please do not deploy the code to any public servers. We do not ensure the security of the code. Turn on one bash for the backend

cd backend
python backend.py

Turn on another bash for the frontend

npm run start

Then, go to your broswer with the below address (do not forget to specify the ip of your PC or server that deploy GCR):

http://localhost:3000/?GCR_Server_IP=your_server_ip