mit-han-lab / bevfusion

[ICRA'23] BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation
https://bevfusion.mit.edu
Apache License 2.0
2.37k stars 428 forks source link

Why is this fix needed? newest release of mmdetection3d still does not do this #598

Closed kabouzeid closed 7 months ago

kabouzeid commented 9 months ago

https://github.com/mit-han-lab/bevfusion/blob/531bf67dd56192cf1cff7baf2995312efdba1ea9/mmdet3d/models/heads/bbox/transfusion.py#L570-L573

zhijian-liu commented 7 months ago

This is because we are using a different coordinate system than mmdetection3d.