Viditya / ultrafast-lane-detection-tf2

TensorFlow 2 / Lite implementation of Ultra-Fast Structure-Aware Lane Detection

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Ultra-Fast Lane Detection in Tensorflow 2 / Lite

This is a TensorFlow 2 and TensorFlow Lite implementation of the Ultra Fast Structure-aware Deep Lane Detection. The aim of this implementation is to run the network on a Google Edge TPU in an embedded device.

The model code in this repository is taken from the original code and adapted to run with TensorFlow instead of pytorch. The ResNet18 implementation used is calmisential's TF2.0 ResNet.

About

TensorFlow 2 / Lite implementation of Ultra-Fast Structure-Aware Lane Detection


Languages

Language:Jupyter Notebook 99.1%Language:Python 0.9%