Framework for Easily Invertible Architectures

[AAAI 2026] The Official Implementation for "Anomagic: Crossmodal Prompt-driven Zero-shot Anomaly Generation"

Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

[arXiv2025] AnoRefiner: Anomaly-Aware Group-Wise Refinement for Zero-Shot Industrial Anomaly Detection. Paper is avaliable at https://arxiv.org/abs/2511.22595

(ICCV 2025) DictAS: A Framework for Class-Generalizable Few-Shot Anomaly Segmentation via Dictionary Lookup

[arXiv2025] MuSc-V2: Zero-Shot Multimodal Industrial Anomaly Classification and Segmentation with Mutual Scoring of Unlabeled Samples. Paper is avaliable at https://arxiv.org/abs/2511.10047

OpenMMLab Pre-training Toolbox and Benchmark

Official repository for CVPR2022 publication, ViM: Out-Of-Distribution with Virtual-logit Matching

Out-of-distribution detection, robustness, and generalization resources. The repository contains a curated list of papers, tutorials, books, videos, articles and open-source libraries etc

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

[ECCV2022] Masked Autoencoders for Point Cloud Self-supervised Learning

PyTorch code and models for the DINOv2 self-supervised learning method.

[ICCV-2021] TransReID: Transformer-based Object Re-Identification

Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"

Official PyTorch implementation of ICCV'25 paper "Fine-grained Abnormality Prompt Learning for Zero-shot Anomaly Detection".

Pytorch implementation of MeanFlow on ImageNet and CIFAR10

[NeurIPS 2025] Pixel-Perfect Depth

Accompanying code for the ICCV2025 paper "MultiADS: Defect-aware Supervision for Multi-type Anomaly Detection and Segmentation in Zero-Shot Learning"

[PR] How to Reduce Change Detection to Semantic Segmentation

[CVPR2025] Main Element Binarization (MEBin) approach introduced in AnomalyNCD

(ICCV'25) Information-Bottleneck Driven Binary Neural Network for Change Detection

[NeurIPS 2021] You Only Look at One Sequence

Code release for "SegLLM: Multi-round Reasoning Segmentation"

YOLOE: Real-Time Seeing Anything [ICCV 2025]

LSNet: See Large, Focus Small [CVPR 2025]

We propose IAD-R1, a universal post-training framework that enhances Vision-Language Models for industrial anomaly detection through a two-stage training strategy.

Reference PyTorch implementation and models for DINOv3

DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding

Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.