search

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

Question about your graph card, memory size and trainiing time for the base model? #7

Closed sinsin1998 closed 1 year ago

sinsin1998 commented 1 year ago

Could you provide the information above? Such that i can tell whether if can run on my machine? Thanks a lot..

SxJyJay commented 1 year ago

Our method trains for two stages. In the first stage, we train a pure lidar-only backbone exactly the same as TransFusion-L. In the second stage, we add the camera branch for multi-modal training. The whole training process can be done on 8 RTX 3090 (24G mem) GPUs. The whole training procedure costs about 3-4 days.

SxJyJay commented 1 year ago

Code has been released, you can try to run it on your machine.