search

hailanyi / TED

Transformation-Equivariant 3D Object Detection for Autonomous Driving
https://arxiv.org/abs/2211.11962
Apache License 2.0
139 stars 32 forks source link

Questions about the TED on the KITTI test set #24

Closed jsc122 closed 1 year ago

jsc122 commented 1 year ago

TED is really a great job, but I have a small problem.

That is, " training data " in " These models are not suitable to directly report results on KITTI test set, please use slightly lower score threshold and train the models on all or 80% training data to achieve a desirable performance on KITTI test set. " refers to the KITTI training set (7481 samples) ? Or the train split (3712 samples)?

Looking forward to your reply!

hailanyi commented 1 year ago

All training set (7481 samples ).

jsc122 commented 1 year ago

All training set (7481 samples ).

Thank you for your reply.