TuSimple / tusimple-benchmark

Download Datasets and Ground Truths: https://github.com/TuSimple/tusimple-benchmark/issues/3
Apache License 2.0
566 stars 180 forks source link

tu simple evaluation resizing #10

Open Msabih opened 5 years ago

Msabih commented 5 years ago

Hi,

The tusimple evaluation takes as input h samples and lane points with the size of 720 x 1280 however if we train a network on smaller size then there are two ways to go about it.

a) resize the network output image to 720 x 1280.

b) resize the points to smaller size and generate y samples/lane points pair.

Which one do you think is best ?