How is the Backbone for Extracting img Features Optimized ？

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

How is the Backbone for Extracting img Features Optimized ？ #24

Open liyih opened 9 months ago

liyih commented 9 months ago

In the paper, you cat the pcd and img_feat into batch_fg_pcd_cams, and then voxelize batch_fg_pcd_cams. But voxelization will loss grad, so how does the parameter in the img_backbone be optimized? Hope for your response! Best