YOLO V8

Ultralytics YOLOv8 is a high-speed, SOTA framework for real-time object detection, segmentation, and pose estimation.

YOLOv8 delivers a unified framework for computer vision tasks (detection, instance segmentation, and image classification) with significant performance gains over its predecessors. Built by Ultralytics, it features a new backbone architecture and an anchor-free head that pushes the mAP on COCO benchmarks while maintaining low latency. Developers can deploy models across diverse environments (from edge devices to cloud GPUs) using a streamlined Python API or CLI. Whether you are tracking objects in 4K video or running pose estimation on a Raspberry Pi, YOLOv8 provides the precision and speed required for production-grade AI.

https://github.com/ultralytics/ultralytics