Awesome GitHub RepositoriesImage Recognition Systems

Darknet is a low-level neural network engine and framework written in C. It is designed for training and deploying deep learning models, with a primary focus on convolutional neural networks. The project serves as a CUDA accelerated deep learning library that offloads heavy mathematical operations to NVIDIA graphics hardware. This acceleration is used to increase processing speed and reduce execution time during the training of large networks. The engine supports a range of activities including deep learning research, image recognition development, and the training of convolutional neural ne

This project is an automated machine learning framework and toolkit designed for training and tuning custom models for classification, regression, and recommendations. It functions as a multimodal machine learning toolkit capable of processing and training models using a combination of text, image, audio, and sensor data. The framework distinguishes itself as a multimodal data processor that can handle and visualize large datasets on a single machine using column-oriented disk storage. It includes a core machine learning model generator that converts trained models into formats compatible wit

ConvNetJS is a JavaScript deep learning library and neural network training engine designed for client-side machine learning. It functions as a framework for building, training, and running convolutional neural networks directly within a web browser without the need for a backend server. The library specializes in image recognition and pattern analysis using convolutional and pooling layers. It enables the creation of models for classification and regression tasks, as well as the development of reinforcement learning agents that optimize behavior through trial and error in simulated environme

KoboldCPP is a local large language model inference engine and GGUF model runner designed to execute quantized models on personal hardware. It functions as a multimodal AI server and API gateway, providing OpenAI-compatible endpoints that allow third-party clients to interact with locally hosted models. The project distinguishes itself as an AI storytelling backend, featuring dedicated tools for long-form narrative management through persistent memory, world lore tracking, and character state management. It further extends its capabilities as a multimodal server capable of processing text, im

This project is a Python implementation of the Faster R-CNN object detection framework. It serves as a convolutional neural network library and tool for locating and classifying multiple objects within images. The framework provides a pre-trained model implementation that allows for object detection inference without manual training. It supports the full lifecycle of object detection, including training detectors on visual datasets to identify and bound specific object classes. The system covers capabilities for computer vision model evaluation, neural network optimization to reduce model si

Anti-Anti-Spider is an automated web scraping toolkit and CAPTCHA bypass framework. It uses convolutional neural networks to recognize characters and digits in image-based security challenges, enabling programmatic access to protected web content. The project functions as an image recognition model trainer, providing a workflow to preprocess labeled image datasets and train custom neural networks. Users can configure model architectures and hyperparameters to align the recognition system with the visual style of specific target websites. The toolkit covers capabilities for image data preproc

The TensorFlow Cookbook is a collection of code examples and recipes for building, training, and deploying machine learning models using TensorFlow. It covers the full model lifecycle, from constructing neural networks and training them with configurable parameters to packaging trained models for production deployment with unit tests and multi-device support. The project also integrates TensorBoard for logging and visualizing computational graphs, scalar summaries, and histograms during training. The cookbook demonstrates a wide range of machine learning techniques, including convolutional ne

is-thirteen es una biblioteca de validación de números y comprobador de igualdad numérica diseñada para verificar si una entrada dada es igual al valor trece. Funciona como una herramienta de clasificación de datos que identifica este valor específico a través de flujos de entrada numéricos, textuales y visuales. El proyecto incluye un clasificador de números basado en imágenes que utiliza aprendizaje profundo y análisis de redes neuronales para identificar representaciones visuales del número trece dentro de imágenes cargadas. La biblioteca cubre una variedad de métodos de validación, incluyendo igualdad aritmética exacta, coincidencia de valores aproximados dentro de rangos de tolerancia definidos, análisis de notación científica y coincidencia de patrones lingüísticos para formas escritas del número.

This is an open-source automation tool for the game Wuthering Waves that uses image recognition to control gameplay without modifying game memory or files. It runs automation tasks while the game window is minimized or obscured, freeing the computer for other use, and accepts command-line arguments to start specific tasks and optionally exit after completion. The tool automatically detects playable characters through screen analysis and adapts actions without manual skill configuration. It supports all common 16:9 resolutions up to 4K as well as some ultrawide formats, with a minimum required

PyBoy es un emulador de Game Boy programable y un framework de simulación de hardware escrito en Python. Funciona como un motor de emulación que permite a los usuarios ejecutar software original de la consola portátil mientras proporciona una interfaz programática para controlar, sondear y automatizar la ejecución de juegos. El proyecto está diseñado específicamente como un entorno de aprendizaje por refuerzo, exponiendo estados y controles del emulador para facilitar el entrenamiento de agentes de machine learning. Se distingue por proporcionar herramientas para el mapeo de áreas de juego y la extracción de representaciones simplificadas de pantalla 2D y mapas de colisión para apoyar la inteligencia artificial. El sistema cubre una amplia gama de capacidades, incluyendo emulación de hardware precisa por ciclo, operaciones de lectura y escritura directa en memoria y un sistema de callbacks para hooks de ejecución. Admite la extracción de datos de juego en tiempo real, como posiciones de sprites y símbolos de memoria, e incluye un modo de ejecución headless para acelerar la velocidad de simulación al omitir el renderizado de gráficos y audio. El emulador también proporciona utilidades para la persistencia de estado mediante serialización de snapshots, simulación de entrada para agentes autónomos y herramientas para el análisis de memoria y modificación de datos de ROM.

Este proyecto es una implementación en PyTorch de la arquitectura Faster R-CNN para la detección de objetos. Proporciona un framework para identificar múltiples clases de objetos y sus cajas delimitadoras (bounding boxes) correspondientes dentro de imágenes utilizando un sistema de aprendizaje profundo. La implementación incluye un pipeline de entrenamiento para optimizar modelos en datasets personalizados y una utilidad para convertir pesos preentrenados de formatos externos a una estructura compatible para la inicialización del modelo. El sistema cubre un pipeline de detección de dos etapas que comprende una red de propuesta de regiones y una capa de pooling ROI. Incorpora funciones de pérdida multitarea y regresión de cajas delimitadoras basada en anclas para refinar las ubicaciones de los objetos. El proyecto incluye herramientas para la visualización en tiempo real de la pérdida de entrenamiento y la precisión de predicción para monitorear el rendimiento del modelo.

Recognize-anything is a multimodal foundation model designed for image recognition, visual tagging, and the generation of descriptive text captions from visual input. It functions as a multimodal embedding model that maps images and text into a shared vector space to enable cross-modal retrieval and recognition. The system implements zero-shot image classification and open-vocabulary object detection, allowing it to recognize object categories not present in the original training data through custom label embeddings. It also features a visual tagging engine and a captioning system that produc