An-Ultra-LightWeight-Lane-Detetctor
the project use PINet as lane detector, supporting training on [VIL-100](https://github.com/yujun0-0/MMA-Net/tree/main/dataset). For the trade-off between the accuracy of lane detection and the inference delay of the model, the project proposes an ultra-lightweight lane line detection method based on a bird's-eye view.The example model's Params and Macs in the project is just 125KB and 66.1MB.