Tags: #computer-vision

A unified framework providing state-of-the-art machine learning models for text, vision, audio, and multimodal tasks, optimized for both inference and training.

Automates complex browser-based workflows using LLMs and computer vision, providing a resilient and adaptive solution for web interaction.

An AI project extending Large Language Models with video understanding capabilities, enabling conversational AI to process and respond to queries about video content.

A comprehensive repository offering C++ and Python code examples for computer vision, deep learning, and AI research, complementing articles on LearnOpenCV.com.

An ultra-compact 7B parameter AI agent designed by Microsoft to automate multi-step computer tasks through visual perception and direct interface interaction.

Enables fast and efficient multi-concept customization of text-to-image diffusion models like Stable Diffusion using a few images.

A PyTorch-based framework to adapt Meta AI's Segment Anything Model (SAM) for improved performance on challenging downstream computer vision tasks using adapters and prompts.

An all-in-one open-source platform for multimodal data labeling and annotation, supporting 3D LiDAR, image, and LLM training data with AI-fueled tools.

A prompt rewriting tool that refines user prompts into clearer, structured versions to enhance the quality of text-to-image generation and image-to-image editing.

A series of video foundation models and large-scale datasets designed for comprehensive multimodal video understanding and generation.

A high-performance, comprehensive face recognition library built on PaddlePaddle and PyTorch.

A stable and generalizable R1-style Large Vision-Language Model (VLM) framework that enhances visual understanding tasks through reinforced learning, outperforming SFT models in generalization.

A comprehensive curated list of resources, papers, datasets, and projects focused on text-to-image generation and manipulation.

A comprehensive research initiative by Microsoft focusing on large-scale self-supervised pre-training to develop advanced foundation models across diverse tasks, languages, and modalities.

Autodistill automates the process of training small, fast supervised models from unlabeled images by leveraging large foundation models, eliminating the need for manual data labeling.

A family of powerful multi-modal large language models (MLLMs) designed to advance AI's understanding and generation capabilities across various data types.

A lightweight C++ toolkit for deploying over 100 diverse AI models with multiple inference engines.

A meticulously categorized collection of research papers on diffusion models, organized by diverse subareas such as visual illusion, color in generation, image restoration, and text-guided editing.

SUPIR is an AI-driven project focused on developing practical algorithms for photo-realistic image restoration and upscaling in real-world scenarios.

A pioneering open-source multimodal AI model family designed to serve as a high-performance alternative to commercial models like GPT-4o and GPT-5.