Technology

YOLOv8-seg

Ultralytics YOLOv8-seg provides high-speed instance segmentation by pinpointing object boundaries with pixel-level precision.

This model architecture from Ultralytics executes simultaneous object detection and mask generation for 80 COCO classes. The YOLOv8n-seg variant delivers 30.5 mAP-mask at 1.1ms inference speeds (A100 TensorRT), balancing accuracy with extreme throughput. It serves critical roles in autonomous navigation (obstacle avoidance), agriculture (crop health monitoring), and robotics (pick-and-place operations). Users can export weights to ONNX or OpenVINO for deployment on diverse hardware: from cloud GPUs to mobile CPUs.

https://docs.ultralytics.com/tasks/segment/

What builders pair with YOLOv8-seg

Projects using both technologies. Select a pairing to see a project.

Built with YOLOv8-seg

Autoencoder 1 shared project Computer Vision 1 shared project Deep learning 1 shared project Detectron2 1 shared project Gaussian Mixture Model 1 shared project Image Processing 1 shared project

12 more pairings

Instance segmentation 1 shared project Isolation Forest 1 shared project K-Nearest Neighbors 1 shared project Local Outlier Factor 1 shared project Mask2Former 1 shared project Mask R-CNN 1 shared project MediaPipe 1 shared project MMDetection 1 shared project One-Class SVM 1 shared project Panoptic FPN 1 shared project Pose tracking 1 shared project Principal Component Analysis 1 shared project

Pairing: Autoencoder

Photo from the event — Event photo

Exercise Posture Tracking Using Computer Vision

Raleigh · February 11, 2026

view project | view Autoencoder

Recent Talks & Demos

Showing 1-1 of 1

Members-Only

Sign in to see who built these projects

Sign in View FAQ

Deadlift Back Curvature Tracking

MediaPipe Pose tracking