TensorRT系列教程之 Windows10下yolov8 tensorrt模型加速部署
https://blog.csdn.net/m0_72734364/article/details/128865904
TensorRT系列教程之 Linux下 yolov8 tensorrt模型加速部署
https://blog.csdn.net/m0_72734364/article/details/128758544
TensorRT系列教程之 Linux下 yolov7 tensorrt模型加速部署
https://blog.csdn.net/m0_72734364/article/details/128759430
TensorRT系列教程之 Linux下 yolov6 tensorrt模型加速部署
https://blog.csdn.net/m0_72734364/article/details/133771238
TensorRT系列教程之 Linux下 yolov5 tensorrt模型加速部署
https://blog.csdn.net/m0_72734364/article/details/133773842
TensorRT系列教程之 Linux下 yolox tensorrt模型加速部署
https://blog.csdn.net/m0_72734364/article/details/133851320
TensorRT系列教程之 Linux下 u2net tensorrt模型加速部署
https://blog.csdn.net/m0_72734364/article/details/133887663?spm=1001.2014.3001.5501
更多:
https://blog.csdn.net/m0_72734364/article/details/133901479?spm=1001.2014.3001.5501
仓库(已更新源码):
https://github.com/FeiYull/tensorrt-alphahttps://github.com/FeiYull/tensorrt-alpha
基于cuda实现模型前处理,后处理,模型从头到脚运行在gpu上。
yolov8和TensorRT-Alpha精度对齐效果:
yolov8n : Offical( left ) vs Ours( right )
基于tensorrt+cuda实现yolov8前处理、后处理加速: