ms eff gcvit deepfake b5 ff plus plus

ProviderKoreaPeter
Categoryvideo-classification
Licensemit
Downloads10.5K
Stars0

Overview

The MS-EFF-GCVit-Deepfake-B5-FF++ is a specialized video classification model designed for high-precision deepfake detection. Unlike general-purpose vision transformers, this model is optimized for identifying synthetic facial manipulations and temporal inconsistencies across video frames. For developers building security layers or content verification pipelines, it offers a robust tool for automating the detection of AI-generated media. The model is released under the MIT license, ensuring seamless integration into both open-source and commercial production environments without restrictive licensing overhead. It serves as a performant alternative to standard CNN-based detectors by leveraging a more sophisticated attention mechanism to spot subtle artifacts in forged videos.

Highlights

  • Specialized in high-accuracy synthetic video and deepfake detection
  • Optimized for temporal inconsistency and facial artifact analysis
  • Permissive MIT license for flexible commercial integration
  • Efficient video classification architecture for production pipelines

Usage

Install
# Install Hugging Face transformers
pip install transformers torch
SDK Usage
# Load model with transformers
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus")
tokenizer = AutoTokenizer.from_pretrained("KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus")

Hugging Face Download

We recommend downloading the model via the Hugging Face CLI or Hub SDK.

Guidance：Before downloading, install huggingface_hub with:

Guidance 
pip install -U huggingface_hub

CLI Download

Download the full repository

Download the full repository 
huggingface-cli download KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus

Download a single file to a local folder (e.g. config.json into ./dir)

Download a single file to a local folder (e.g. config.json into ./dir) 
huggingface-cli download KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus config.json --local-dir ./dir

See the official docs for more CLI options

SDK Download

SDK Download 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus')

Git Download

Make sure git-lfs is installed first

Git Download 
git lfs install
git clone https://huggingface.co/KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus

To skip LFS large-file downloads, use:

Skip LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus

Model files are hosted on the Hugging Face Hub — download directly via HF CLI / SDK / Git, not through this site.

PyTorch / Transformers Usage

Install Transformers

Install Transformers 
pip install -U transformers torch

Load the model and run inference

Load the model and run inference 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus')
tokenizer = AutoTokenizer.from_pretrained('KoreaPeter/ms-eff-gcvit-deepfake-b5-ff-plus-plus')
View official source
Join our Telegram