A collection of 419 pre-trained AI models covering vision, speech, NLP and LLMs, all runnable from a shared Python CLI via the ailia SDK.
2,4 тыс.367PythonOtherОбновлён 1 час назад
sherlockchou86/VideoPipe
A C++ video analysis framework that chains stream reading, model inference, tracking and output as pluggable pipeline nodes.
Open source альтернатива:NVIDIA DeepStream
- Звёзды
- 2,984
- Форки
- 470
- Открытые issues
- 4
- Последний коммит
- 8 месяцев назад
Что это?
VideoPipe is a C++ framework for video analysis and structuring. It is built around a pipeline of independent nodes that can be combined in different ways, so developers can assemble applications such as face recognition, image search, object tracking and traffic or security behavior analysis without writing the plumbing themselves.
The framework handles stream input over protocols like RTSP, RTMP, UDP and files, decoding and encoding through OpenCV or GStreamer with hardware acceleration, and on-screen display of results. Inference uses OpenCV::DNN by default and can be switched to TensorRT, Paddle Inference, ONNX Runtime or a custom backend. Recent versions also integrate multimodal LLMs served through Ollama, vLLM or OpenAI-compatible APIs.
VideoPipe supports multi-level inference, IOU and SORT tracking, line-crossing and parking detection, and pushing structured data to Kafka or sockets. It builds on CMake with optional CUDA, TensorRT, Paddle, Kafka and LLM modules, and runs on CPU when those are disabled.
Ключевые возможности
- 01Reads streams over RTSP, RTMP, UDP, files and applications.
- 02Supports pluggable inference backends including OpenCV::DNN, TensorRT, Paddle Inference and ONNX Runtime.
- 03Performs multi-level inference for detection, classification and feature extraction.
- 04Tracks objects with IOU and SORT and analyzes line-crossing and parking behavior.
- 05Integrates multimodal LLMs via Ollama, vLLM or OpenAI-compatible APIs.
- 06Pushes structured data to Kafka or sockets and overlays results on frames.
Сценарии использования
- Building traffic incident or violation detection from camera feeds.
- Prototyping face recognition or vehicle tracking on edge devices such as Jetson.
- Running video structuring pipelines on CPU-only machines for testing.
- Sending detected events to Kafka for downstream processing.
Темы
Частые вопросы
- Что такое VideoPipe?
- A C++ video analysis framework that chains stream reading, model inference, tracking and output as pluggable pipeline nodes. VideoPipe is a C++ framework for video analysis and structuring. It is built around a pipeline of independent nodes that can be combined in different ways, so developers can assemble applications such as face recognition, image search, object tracking and traffic or security behavior analysis without writing the plumbing themselves.
- VideoPipe можно использовать бесплатно? Какая у него лицензия?
- VideoPipe распространяется с открытым исходным кодом под лицензией Apache-2.0. Перед использованием в коммерческих продуктах проверьте условия лицензии.
- VideoPipe всё ещё поддерживается?
- Нет коммитов больше 6 месяцев. Последний коммит в VideoPipe — 25 февр. 2026 г..
- Для чего можно использовать VideoPipe?
- Building traffic incident or violation detection from camera feeds. Prototyping face recognition or vehicle tracking on edge devices such as Jetson. Running video structuring pipelines on CPU-only machines for testing. Sending detected events to Kafka for downstream processing.
- Где скачать VideoPipe?
- Исходный код и инструкции по установке — на GitHub: https://github.com/sherlockchou86/VideoPipe
Описание создано ИИ на основе README проекта. Статистика GitHub обновляется ежедневно.
Похожие проекты
Инструмент на Python для восстановления лиц вслепую: улучшает размытые, низкокачественные и повреждённые лица на фото и видео с помощью трансформера с поиском по кодовой книге.
18,2 тыс.3,7 тыс.PythonOtherОбновлён 11 месяцев назад
YuE2 — открытая модель генерации музыки: она создает редактируемый план мелодии и аккордов по тексту песни и описанию стиля, а затем превращает его в полноценную песню с вокалом.
ИИ и машинное обучение#music-generation#audio-generation#deep-learning#text-to-audio#music-transcription
12,3 тыс.1,4 тыс.PythonApache-2.0Обновлён 13 дней назад
Open-source Python OCR library that detects and recognizes text in PDFs and images using deep learning models, built on PyTorch.
6,4 тыс.687PythonApache-2.0Обновлён 4 часа назад
Converts static diagram images into editable DrawIO (XML) files using SAM 3 segmentation, OCR, and LaTeX formula recognition.
5,5 тыс.353PythonAGPL-3.0Обновлён позавчера
Converts UI screenshots and design mockups into editable HTML/CSS using a modular multi-agent pipeline with multiple model backends.
3 тыс.320PythonApache-2.0Обновлён в прошлом месяце