HybridNets: End-to-End Perception Network
-
Updated
Dec 1, 2023 - Python
HybridNets: End-to-End Perception Network
使用OpenCV部署HybridNets,同时处理车辆检测、可驾驶区域分割、车道线分割,三项视觉感知任务,包含C++和Python两种版本的程序实现。本套程序只依赖opencv库就可以运行, 彻底摆脱对任何深度学习框架的依赖。
Offline AI dashcam analysis for road perception, risk events, and privacy-first telematics research.
Advanced autonomous driving perception suite comparing classical computer vision and deep learning based lane/road detection pipelines, including HybridNets ONNX multitask inference, U-Net road segmentation, curved lane tracking, and real-time GPU-accelerated autonomous road analysis.
To associate your repository with the hybridnets topic, visit your repo's landing page and select "manage topics."