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
181 stars 11 forks source link

提交的问题 #28

Open rubbish001 opened 10 months ago

rubbish001 commented 10 months ago

我弄了一个融合模型，我在验证集上跑的效果是mAP0.696，nds是0.731，但是提交的效果很差，mAp是0.64，一般是什么原因导致的，提交的格式有改变吗

SxJyJay commented 10 months ago

For leaderboard submission, you should merge training and validation sets to re-train your model.

rubbish001 commented 10 months ago

你用的啥主干网络跑的0.715，我用的是v299

SxJyJay commented 10 months ago

We use ResNet-50 as image backbone

Libraaer commented 3 weeks ago

你用的啥主干网络跑的0.715，我用的是v299

Hello, have you trained this model? I'm using A800 here but still not enough to train. If you've trained, can you give me an answer to the reason for my lack of memory here, thank you