How can I train my own images for detecting new objects?

[jinj]

DriveNet is very good, but I want to recognize more objects.

[SteveNV]

Dear jin_j,

Please refer to below link for your topic. Thanks.

https://devtalk.nvidia.com/default/topic/1031112/driveworks/how-to-improve-and-train-drivenet-or-lanenet-in-driveworks-/