Object detection

Supervision is a Python toolkit for building computer vision applications, providing reusable components for model integration, real-time annotation,...

React Native ExecuTorch enables on-device AI model inference in React Native applications using Meta's ExecuTorch framework, supporting computer...

RF-DETR is a real-time transformer-based architecture for object detection, instance segmentation, and keypoint detection, built on a DINOv2 backbone...

Inference is a production-ready platform that deploys computer vision models to edge devices and servers, enabling real-time inference with support...

X-AnyLabeling is a desktop application for image and video annotation with integrated AI models (SAM, YOLO, GroundingDINO, OCR, etc.) that automates...

Viseron is a self-hosted, local-only NVR (Network Video Recorder) with AI computer vision capabilities including object detection, motion detection,...

Geti is an end-to-end Vision AI application that accelerates computer vision model development from raw images to deployed inference, optimized for...

D-FINE is a specialized DETR-based object detector that reformulates bounding box regression as fine-grained distribution refinement with...

DeepCamera is an open-source AI camera platform that enables local video analysis using Vision Language Models (Qwen, DeepSeek, LLaVA) and object...

PaddleX is a low-code AI development tool built on PaddlePaddle that integrates 270+ pre-trained models across document analysis, OCR, object...

PyTorch-GradCAM is a specialized library providing state-of-the-art explainability methods for computer vision models, enabling visual interpretation...

Ultralytics YOLO is a Python library providing state-of-the-art real-time computer vision models including YOLO11, YOLOv8, and YOLO26 for object...

LightlyTrain is a specialized framework for computer vision model development, covering pretraining (DINOv2/v3), fine-tuning (YOLO, ViTs, RT-DETR),...

Frigate is a self-hosted NVR (Network Video Recorder) with real-time local object detection for IP cameras, designed specifically for Home Assistant...

LocalAI is an open-source engine that runs any AI model (LLMs, vision, audio, image, video) on any hardware without GPU requirements, using a...

CVAT is a production-grade, open-source platform for annotating images, videos, and 3D data with AI-assisted labeling, team collaboration, and...

A high-performance TensorFlow Lite library for React Native that provides GPU-accelerated ML inference via Nitro Modules with zero-copy ArrayBuffers....

The ZED SDK is a cross-platform C++ library for stereo vision and depth-perception hardware (ZED cameras), enabling real-time 3D reconstruction,...

SAHI is a lightweight computer vision library specialized in performing object detection and instance segmentation on large images through...

FiftyOne is an open-source platform for building, curating, and evaluating computer vision datasets and models through interactive visualization,...

camera.ui is a self-hosted video surveillance platform offering live viewing, local AI detection, and smart-home integration without mandatory cloud...

MLKit is a Kotlin wrapper around Google's ML Kit vision APIs that simplifies computer vision tasks on Android, including barcode scanning, OCR, face...

A collection of 61+ Jupyter notebooks providing tutorials on state-of-the-art computer vision models including YOLOv11, SAM 2, Florence-2, PaliGemma,...

CoreML-Models is a curated collection of pre-converted machine learning models optimized for Apple's CoreML framework, enabling developers to...

AutoGluon is an automated machine learning framework developed by AWS that enables users to train high-accuracy ML and deep learning models across...

NNCF is a specialized neural network compression framework for optimizing inference performance in OpenVINO and other backends (PyTorch, ONNX)...

This is a Home Assistant integration for Frigate, an open-source NVR with AI object detection capabilities. It bridges Frigate and Home Assistant by...

DeepDetect is a deep learning runtime and REST server for training and inference across multiple frameworks (PyTorch, TensorRT, NCNN, XGBoost). It...

Ultralytics YOLOv5 is a PyTorch-based real-time object detection model that also supports image segmentation and classification, with export paths to...

torchdistill is a configuration-driven framework for knowledge distillation and reproducible deep learning experiments built on PyTorch. It enables...