3D Object Detection on nuScenes (val) (Primary & Error Metrics) benchmark leaderboard · SOTA2 Research

3D Object Detection on nuScenes (val) (Primary & Error Metrics)

51.7NDSMambaBEV-B

Evaluation Results

Method	Links
MambaBEV-B2024.10		51.7	42.7	0.669	0.265	0.365	0.468	0.193
TempBEV2024.10		50.8	40.8	—	—	—	—	—
EPro-PnP-Detv22024.10		49	42.3	0.547	0.236	0.302	1.071	0.123
PolarDETR-T2024.10		48.8	38.3	0.707	0.269	0.344	0.518	0.196
UVTR2024.10		48.3	37.9	0.731	0.267	0.35	0.51	0.2
DD3D2024.10		47.7	41.8	0.572	0.249	0.368	1.014	0.124
Ego3RT2024.10		47.3	42.5	0.55	0.264	0.433	1.014	0.145
BEVFormer-s2024.10		44.8	37.5	0.725	0.272	0.391	0.802	0.2
PolarDETR2024.10		44.4	36.5	0.742	0.269	0.35	0.829	0.197
PETR2024.10		44.2	37	0.711	0.267	0.383	0.865	0.201
DETR3D2024.10		42.5	34.6	0.773	0.268	0.383	0.842	0.216
PGD2024.10		40.9	33.5	0.732	0.263	0.423	1.285	0.172
BEVDet2024.10		38.4	31.7	0.704	0.273	0.531	0.94	0.25
FCOS3D2024.10		37.2	29.5	0.806	0.268	0.511	1.315	0.17
MambaBEV-T2024.10		36.8	26.2	0.881	0.2909	0.5998	0.6366	0.2209
BEVFormer-T2024.10		35.4	25.2	0.9	0.294	0.655	0.657	0.216