lucastabelini / PolyLaneNet

Code for the paper entitled "PolyLaneNet: Lane Estimation via Deep Polynomial Regression" (ICPR 2020)
https://arxiv.org/abs/2004.10924
MIT License
295 stars 74 forks source link

a question, :-) #1

Closed lzcchl closed 4 years ago

lzcchl commented 4 years ago

Hi, gratefully to share your project, and your models are so much, in what case run 115FPS, which backbone? what intput size? is that tusimple_320x180?

lucastabelini commented 4 years ago

The full model runs at 115 FPS on a Titan V. It uses EfficientNet-b0 as backbone with an input size of 640x360. The models tusimple and tusimple_fulltrain use those settings, the only difference is that the latter uses the full training set (train+val).