All-in-one training for vision models (YOLO, ViTs, RT-DETR, DINOv3): pretraining, fine-tuning, distillation.

A microframework on top of PyTorch with first-class citizen APIs for foundation model adaptation

[ICLR'24 & IJCV‘25] Matcher: Segment Anything with One Shot Using All-Purpose Feature Matching

Testing adaptation of the DINOv2/3 encoders for vision tasks with Low-Rank Adaptation (LoRA)

[ECCV 2024] Improving 2D Feature Representations by 3D-Aware Fine-Tuning

Unofficial implementation of the paper "The Chosen One: Consistent Characters in Text-to-Image Diffusion Models"

[WACV2025] AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2

[CVPR'24] NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild

[ICCV 2025] Official repository of the paper "Talking to DINO: Bridging Self-Supervised Vision Backbones with Language for Open-Vocabulary Segmentation"

Welcome to the project repository for POPE (Promptable Pose Estimation), a state-of-the-art technique for 6-DoF pose estimation of any object in any scene using a single reference.

Integrating SAM2 with DINOv2/v3 for segmentation

A cli program of image retrieval using dinov2

[NeurIPS'24] A Simple Image Segmentation Framework via In-Context Examples

[WACV 2026] Official implementation of the paper: “CountingDINO: A Training-free Pipeline for Exemplar-based Class-Agnostic Counting”

Official implementation of the paper 'Exploring Robust Features for Few-Shot Object Detection in Satellite Imagery'

[ICLR 2026] The implementation of the paper Foundation Visual Encoders Are Secretly Few-Shot Anomaly Detectors

This project is an image retrieval system based on DINOv2 and CLIP models. It uses Chroma vector database to support both text-to-image and image-to-image retrieval.

Code for the paper "Robust representations for image classification via counterfactual contrastive learning" (Medical Image Analysis) and "Counterfactual contrastive learning: robust representations via causal image synthesis" (MICCAI Data Engineering Workshop)

This repo contains the official implementation of ICCV 2025 paper "MoSiC: Optimal-Transport Motion Trajectory for Dense Self-Supervised Learning

The inference of DINOv2 ONNX models using the ONNXRuntime library.

Vision Foundation Models: SAM, ViT, CLIP, DINOv2, object detection, segmentation, and multimodal AI for computer vision.

Code for MM-DINOv2: Adapting Foundation Models for Multi-Modal Medical Image Analysis (MICCAI2025)

Dinov3 exploration (use cases & capabilities )

[SIGGRAPH Asia 2025] Official implementation of Surface-Aware Distilled 3D Semantic Features

[ICML 2025] Official implementation of SPEC method for interpretable embedding comparison. paper: Towards an Explainable Comparison and Alignment of Feature Embeddings

DINOv2 module for use with Autodistill.

An open-source implementaion for fine-tuning DINOv2 by Meta.

[HAIS 2025] MapFM: Foundation Model-Driven HD Mapping with Multi-Task Contextual Learning

This project implements knowledge distillation from DINOv2 (Vision Transformer) to convolutional networks, enabling efficient visual representation learning with reduced computational requirements.

Fusion at the Foregut: CLIP-Based Prototypical Learning with DINOv2 Refinement for Endoscopic Image Analysis