emptysoal/TensorRT-YOLOv8

Based on tensorrt v8.0+, deploy detect, pose, segment, tracking of YOLOv8 with C++ and python api.

52
/ 100
Established

This project helps developers integrate high-performance computer vision tasks like object detection, pose estimation, instance segmentation, and object tracking into their applications. It takes pre-trained YOLOv8 models and optimizes them for faster execution using TensorRT, providing a C++ API for maximum speed and a Python API for ease of use. This is ideal for developers building vision-powered systems on edge devices like Jetson or Linux servers.

141 stars.

Use this if you are a software developer needing to deploy YOLOv8 models for real-time object detection, pose estimation, instance segmentation, or tracking with high inference speed on NVIDIA GPUs.

Not ideal if you are an end-user without programming experience or if you need to train YOLOv8 models, as this project focuses on deployment rather than training.

real-time vision edge AI deployment computer vision engineering embedded systems video analytics
No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

141

Forks

27

Language

C++

License

MIT

Last pushed

Oct 24, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/emptysoal/TensorRT-YOLOv8"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.