A curated list of recent diffusion models for video generation, editing, restoration, understanding, etc.

[CVPR 2023 Highlight] Official implementation of the paper: "AltFreezing for More General Video Face Forgery Detection"

Official repo for AIGCBench: Comprehensive Evaluation of Image-to-Video Content Generated by AI

APBench: A Unified Availability Poisoning Attack and Defenses Benchmark

The official PyTorch implementation of the paper "Human Motion Diffusion Model"

🚀 PyTorch Implementation of "Progressive Distillation for Fast Sampling of Diffusion Models(v-diffusion)"

An adversarial attack on object detectors

深度学习经典、新论文逐段精读

[TMLR'24] On the Adversarial Robustness of Camera-based 3D Object Detection

Repository for patch attacks against autonomous driving vision tasks.

Synthetic Image Detection

Flat Seeking Bayesian Neural Networks

Spectrum simulation attack (ECCV'2022 Oral) towards boosting the transferability of adversarial examples

A comprehensive benchmark of deepfake detection

Github of the FaceForensics dataset

Official code for the paper Towards Bridging the Performance Gaps of Joint Energy-based Models

PyTorch implementation of JEM++: Improved Techniques for Training JEM

Code for our ICLR 2023 paper Making Substitute Models More Bayesian Can Enhance Transferability of Adversarial Examples.

[NeurIPS 2023] Boosting Adversarial Transferability by Achieving Flat Local Maxima

EasyRobust: an Easy-to-use library for state-of-the-art Robust Computer Vision Research with PyTorch.

An unofficial implementation of the paper《Towards Understanding and Boosting Adversarial Transferability from a Distribution Perspective》

🔥🔥Defending Against Deepfakes Using Adversarial Attacks on Conditional Image Translation Networks

The official code of IEEE S&P 2024 paper "Why Does Little Robustness Help? A Further Step Towards Understanding Adversarial Transferability". We study how to train surrogates model for boosting tra…

Beyond imagenet attack (accepted by ICLR 2022) towards crafting adversarial examples for black-box domains.

An unrestricted attack based on diffusion models that can achieve both good transferability and imperceptibility.

[NeurIPS 2023] Codes for DiffAttack: Evasion Attacks Against Diffusion-Based Adversarial Purification