SxJyJay / MSMDFusion

[CVPR 2023] MSMDFusion: Fusing LiDAR and Camera at Multiple Scales with Multi-Depth Seeds for 3D Object Detection
Apache License 2.0
167 stars 10 forks source link

About TTA and base version. #12

Open syy-whu opened 1 year ago

syy-whu commented 1 year ago

Hello! I want to ask about the differences between TTA and the base version. And why can the TTA version improve such a large margin?

SxJyJay commented 1 year ago

TTA adopts different data augmentation techniques (scaling and flip) only during testing and ensembles their predictions for stronger performances.