Пайплайны компьютерного зрения с YOLOv8 и Detectron2

★ 6.8 · general

computer-vision-pipeline is a Claude Code skill that builds production-ready computer vision pipelines for object detection, tracking, and video analysis, with support for YOLOv8, Detectron2, TensorFlow, and PyTorch. It covers the full workflow: frame preprocessing, model selection based on speed or accuracy requirements, custom model fine-tuning, and inference. Primary use cases include drone footage analysis for archaeological surveys and conservation, wildlife monitoring, multi-object tracking (MOT), and security systems. The skill is not intended for simple image filters, photo editing tools, or facial recognition APIs such as AWS Rekognition.