search

tusen-ai / SST

Code for a series of work in LiDAR perception, including SST (CVPR 22), FSD (NeurIPS 22), FSD++ (TPAMI 23), FSDv2, and CTRL (ICCV 23, oral).
Apache License 2.0
765 stars 96 forks source link

Could you share the results of CTRL base detector？ #169

Closed JuiceLemonLemon closed 6 months ago

JuiceLemonLemon commented 9 months ago

And the performance of adopted base detector is:

OBJECT_TYPE_TYPE_VEHICLE_LEVEL_1: [mAP 0.82786] [mAPH 0.823132] OBJECT_TYPE_TYPE_VEHICLE_LEVEL_2: [mAP 0.751349] [mAPH 0.746877] OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_1: [mAP 0] [mAPH 0] OBJECT_TYPE_TYPE_PEDESTRIAN_LEVEL_2: [mAP 0] [mAPH 0] OBJECT_TYPE_TYPE_SIGN_LEVEL_1: [mAP 0] [mAPH 0] OBJECT_TYPE_TYPE_SIGN_LEVEL_2: [mAP 0] [mAPH 0] OBJECT_TYPE_TYPE_CYCLIST_LEVEL_1: [mAP 0] [mAPH 0] OBJECT_TYPE_TYPE_CYCLIST_LEVEL_2: [mAP 0] [mAPH 0] RANGE_TYPEVEHICLE[0, 30)_LEVEL_1: [mAP 0.93624] [mAPH 0.932365] RANGE_TYPEVEHICLE[0, 30)_LEVEL_2: [mAP 0.925796] [mAPH 0.921951] RANGE_TYPEVEHICLE[30, 50)_LEVEL_1: [mAP 0.819024] [mAPH 0.813976] RANGE_TYPEVEHICLE[30, 50)_LEVEL_2: [mAP 0.761934] [mAPH 0.757116] RANGE_TYPEVEHICLE[50, +inf)_LEVEL_1: [mAP 0.668922] [mAPH 0.661151] RANGE_TYPEVEHICLE[50, +inf)_LEVEL_2: [mAP 0.542161] [mAPH 0.535483]

@Abyssaledge Hello, Good morning. Above is your shared CTRL base detector results. And could you please share me the results of other classes (ped, cyclist)? and the detector results, I fould you just shared the detector+ImmortalTracker results？

Abyssaledge commented 9 months ago

Sorry for the late reply. I will check it tomorrow.

JuiceLemonLemon commented 8 months ago

@Abyssaledge Hello, could you please share the fsd_base_cyc_trainset.bin?

Abyssaledge commented 6 months ago

Sorry for the late reply, the requested file has been uploaded.