search

ZrrSkywalker / MonoDETR

[ICCV 2023] The first DETR model for monocular 3D object detection with depth-guided transformer
327 stars 31 forks source link

Thank you very much and can you provide the best version for the test set #48

Open BeyondVE opened 7 months ago

BeyondVE commented 7 months ago

as title

Vipermdl commented 6 months ago

我跑出来只有这个指标。 Car (3D Detection) | 19.51 % | 12.70 % | 10.75 % Car (Bird's Eye View) | 29.81 % | 19.31 % | 15.98 %

BeyondVE commented 6 months ago

我跑出来只有这个指标。 Car (3D Detection) | 19.51 % | 12.70 % | 10.75 % Car (Bird's Eye View) | 29.81 % | 19.31 % | 15.98 %

我跑出来的测试集是这样 没达到论文指标 Car AP_R40@0.70, 0.70, 0.70: 3d AP:24.9635, 18.9295, 15.7721

Ivan-Tang-3D commented 5 months ago

Please try to revise the code version to the original version and then run several runs to get the mean value. Please refer to this link https://github.com/ZrrSkywalker/MonoDETR/issues/51.