3D Object Detection on nuScenes (test)
76.5mAPEA-LSS
Evaluation results
| Method | Links | |||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| EA-LSSModality=C+L, Baseline=BEVFusion (Liang et al. 2022), Test time augmentation=true, Model ensemble=true2023.03 | 76.5 | 77.6 | — | — | — | — | — | — | — | — | — | — | 0.233 | 0.228 | 0.281 | 0.196 | 0.123 | — | — | — | — | — | — | — | — | — | — | |
| IS-FUSIONModality=LiDAR + Camera, test-time augmentation=true, model ensemble=true2024.03 | 76.5 | 77.4 | 89.8 | 67.8 | 77.6 | 68.3 | 44.5 | 93.4 | 85.3 | — | 91.1 | — | — | — | — | — | — | — | 65.6 | — | — | 81.8 | — | — | — | — | — | |
| DeepInteraction-eModality=L+C, Image Backbone=Swin-Tiny, LiDAR Backbone=VoxelNet, Test-time augmentation=true, Model ensemble=true2022.08 | 75.6 | 76.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusion-e [30]Modality=L+C, Image Backbone=Swin-Tiny, LiDAR Backbone=VoxelNet, Test-time augmentation=true, Model ensemble=true2022.08 | 75 | 76.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Co-Fusion4DModality=LC2026.05 | 74.9 | 75.6 | 90.5 | 65.7 | 77.1 | — | 39.7 | 91.2 | 83.5 | — | 88.2 | — | — | — | — | — | — | — | 63.1 | — | — | 81 | — | — | 69.9 | — | — | |
| MV2DFusionVenue=TPAMI’25, Modality=LC2026.06 | 74.5 | 76.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepInteraction-largeModality=L+C, Image Backbone=Swin-Tiny, LiDAR Backbone=VoxelNet, Test-time augmentation=true, Model ensemble=false2022.08 | 74.1 | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GAFusionModality=LC2024.11 | 73.6 | 74.9 | 89.4 | 65.3 | 73.7 | 65.8 | 42.4 | 92.3 | 80.8 | — | 87 | 79.2 | — | — | — | — | — | — | 60.2 | — | — | — | — | — | — | — | — | |
| GAFusionModality=LC2026.05 | 73.6 | 74.9 | 89.4 | 65.3 | 73.7 | — | 42.4 | 92.3 | 80.8 | — | 87 | — | — | — | — | — | — | — | 60.2 | — | — | 79.2 | — | — | 65.8 | — | — | |
| BEVFusion4DModality=LC2026.05 | 73.3 | 74.7 | 89.7 | 65.6 | 72.9 | — | 41.1 | 90.9 | 79.5 | — | 87.7 | — | — | — | — | — | — | — | 58.6 | — | — | 81 | — | — | 66 | — | — | |
| IS-FUSIONModality=LiDAR + Camera, test-time augmentation=false, model ensemble=false2024.03 | 73 | 75.2 | 88.3 | 62.7 | 74.9 | 67.3 | 38.4 | 89.3 | 82.4 | — | 89.2 | — | — | — | — | — | — | — | 59.5 | — | — | 78.1 | — | — | — | — | — | |
| IS-FusionModality=LC2026.05 | 73 | 75.2 | 88.3 | 62.7 | 74.9 | — | 38.4 | 89.3 | 82.4 | — | 89.2 | — | — | — | — | — | — | — | 59.5 | — | — | 78.1 | — | — | 67.3 | — | — | |
| CMT-Large + ALIGNQuery Init.=ALIGN, Test-time augmentation=false2025.12 | 72.4 | 74.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Co-Fix3DModality=LC2026.05 | 72.3 | 74.7 | 89.6 | 60.8 | 73.3 | — | 37.9 | 88.7 | 77.6 | — | 86.7 | — | — | — | — | — | — | — | 62.7 | — | — | 80 | — | — | 65.2 | — | — | |
| EA-LSSModality=C+L, Baseline=BEVFusion (Liang et al. 2022)2023.03 | 72.2 | 74.4 | — | — | — | — | — | — | — | — | — | — | 0.247 | 0.237 | 0.304 | 0.25 | 0.133 | — | — | — | — | — | — | — | — | — | — | |
| MEFormerModality=L+C2024.07 | 72.2 | 74.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseFusionModality=LiDAR + Camera2024.03 | 72 | 73.8 | 88 | 60.2 | 72 | 64.9 | 38.7 | 90.9 | 78.5 | — | 87.9 | — | — | — | — | — | — | — | 59.8 | — | — | 79.2 | — | — | — | — | — | |
| CMTModality=LiDAR + Camera2024.03 | 72 | 74.1 | 88 | 63.3 | 75.4 | 65.4 | 37.3 | 87.9 | 79.1 | — | 84.7 | — | — | — | — | — | — | — | 60.6 | — | — | 78.2 | — | — | — | — | — | |
| SparseFusionModality=L+C2024.07 | 72 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CMTModality=L+C2024.07 | 72 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseFusionQuery Init.=Heatmap, Test-time augmentation=false2025.12 | 72 | 73.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CMT-LargeQuery Init.=Random, Test-time augmentation=false2025.12 | 72 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SparseFusionModality=LC2026.05 | 72 | 73.8 | 88 | 60.2 | 72 | — | 38.7 | 90.9 | 78.5 | — | 87.9 | — | — | — | — | — | — | — | 59.8 | — | — | 79.2 | — | — | 64.9 | — | — | |
| SparseFusionVenue=ICCV’23, Modality=LC2026.06 | 72 | 73.8 | 88 | 60.2 | 72 | 64.9 | 38.7 | 90.9 | 78.5 | — | 87.9 | 79.2 | — | — | — | — | — | — | 59.8 | — | — | — | — | — | — | — | — | |
| CMTVenue=ICCV’23, Modality=LC2026.06 | 72 | 73.8 | 88 | 63.3 | 75.4 | 65.4 | 37.3 | 87.9 | 79.1 | — | 84.7 | 78.2 | — | — | — | — | — | — | 60.6 | — | — | — | — | — | — | — | — | |
| GraphBEV++ (LSS)Modality=LC2026.06 | 72 | 74 | 89.5 | 60.5 | 72.4 | 64.9 | 41.3 | 91.1 | 77.1 | — | 89.3 | 80.3 | — | — | — | — | — | — | 53.6 | — | — | — | — | — | — | — | — | |
| MTAModality=LC2026.05 | 71.8 | 73.7 | 88.6 | 63.3 | 71.9 | — | 38.4 | 90.8 | 75.5 | — | 88.5 | — | — | — | — | — | — | — | 57.2 | — | — | 79.4 | — | — | 64.7 | — | — | |
| GraphBEVModality=LC2026.05 | 71.7 | 73.6 | 89.2 | 60 | 72.1 | — | 40.8 | 90.9 | 76.8 | — | 88.9 | — | — | — | — | — | — | — | 53.3 | — | — | 80.1 | — | — | 64.5 | — | — | |
| GraphBEVVenue=ECCV’24, Modality=LC2026.06 | 71.7 | 73.6 | 89.2 | 60 | 72.1 | 64.5 | 40.8 | 90.9 | 76.8 | — | 88.9 | 80.1 | — | — | — | — | — | — | 53.3 | — | — | — | — | — | — | — | — | |
| FocalFormer3D-FDevice=V100 GPU2023.08 | 71.6 | 73.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 363 | — | — | — | — | — | — | |
| FocalFormer3DModality=LiDAR + Camera2024.03 | 71.6 | 73.9 | 88.5 | 61.4 | 71.7 | 66.4 | 35.9 | 89.7 | 80.3 | — | 85.3 | — | — | — | — | — | — | — | 57.1 | — | — | 79.3 | — | — | — | — | — | |
| FocalFormer3DModality=LC2024.11 | 71.6 | 73.9 | 88.5 | 61.4 | 71.7 | 66.4 | 35.9 | 89.7 | 80.3 | — | 85.3 | 79.3 | — | — | — | — | — | — | 57.1 | — | — | — | — | — | — | — | — | |
| FocalFormer3DQuery Init.=Heatmap, Test-time augmentation=false2025.12 | 71.6 | 73.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LiDARMultiNetModality=L2026.05 | 71.6 | 67 | 86.9 | 57.4 | 64.7 | — | 31.5 | 87.2 | 75.3 | — | 85.1 | — | — | — | — | — | — | — | 47.6 | — | — | 73.5 | — | — | 61 | — | — | |
| FocalFormer3DModality=LC2026.05 | 71.6 | 73.9 | 88.5 | 61.4 | 71.7 | — | 35.9 | 89.7 | 80.3 | — | 85.3 | — | — | — | — | — | — | — | 57.1 | — | — | 79.3 | — | — | 66.4 | — | — | |
| MSMDFusionModality=LC, Single-model=true, Ensemble=false, Test-time augmentation=false2022.09 | 71.5 | 74 | 88.4 | 61 | 71.4 | 64.2 | 35.2 | 90.6 | 76.9 | — | 88.1 | 80.7 | — | — | — | — | — | — | 58.3 | — | — | — | — | — | — | — | — | |
| MSMDFusionModality=LiDAR + Camera2024.03 | 71.5 | 74 | 88.4 | 61 | 71.4 | 64.2 | 35.2 | 90.6 | 76.9 | — | 88.1 | — | — | — | — | — | — | — | 58.3 | — | — | 80.7 | — | — | — | — | — | |
| MSMDFusionModality=LC2024.11 | 71.5 | 74 | 88.4 | 61 | 71.4 | 66.2 | 35.2 | 90.6 | 76.9 | — | 88.1 | 80.7 | — | — | — | — | — | — | 58.3 | — | — | — | — | — | — | — | — | |
| MSMDFusionModality=LC2026.05 | 71.5 | 74 | 88.4 | 61 | 71.4 | — | 35.2 | 90.6 | 76.9 | — | 88.1 | — | — | — | — | — | — | — | 58.3 | — | — | 80.7 | — | — | 64.2 | — | — | |
| MSMDFusionVenue=CVPR’23, Modality=LC2026.06 | 71.5 | 74 | 88.4 | 61 | 71.4 | 64.2 | 35.2 | 90.6 | 76.9 | — | 88.1 | 80.7 | — | — | — | — | — | — | 58.3 | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LC, BEV-space data augmentation=true2022.05 | 71.3 | 73.3 | 88.5 | 63.1 | 72 | 64.7 | 38.1 | 90 | 75.2 | — | 86.5 | 78.3 | — | — | — | — | — | — | 56.5 | — | — | — | — | — | — | — | — | |
| BEVFusionModality=C+L2023.03 | 71.3 | 73.3 | — | — | — | — | — | — | — | — | — | — | 0.25 | 0.24 | 0.359 | 0.254 | 0.132 | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=L+C, Citation=[13]2024.07 | 71.3 | 73.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LC, citation=[22]2024.11 | 71.3 | 73.3 | 88.1 | 60.9 | 69.3 | 62.1 | 34.4 | 89.2 | 72.2 | — | 86.7 | 78.2 | — | — | — | — | — | — | 52.2 | — | — | — | — | — | — | — | — | |
| EfficientQ3MQuery Init.=Heatmap, Test-time augmentation=false2025.12 | 71.2 | 73.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LargeKernel3D-FModality=L+C, Image Backbone=R50, LiDAR Backbone=VoxelNet-LargeKernel, Test-time augmentation=true, Model ensemble=false2022.08 | 71.1 | 74.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ObjectFusionModality=LiDAR + Camera2024.03 | 71 | 73.3 | 89.4 | 59 | 71.8 | 63.1 | 40.5 | 90.7 | 78.1 | — | 87.7 | — | — | — | — | — | — | — | 53.2 | — | — | 80 | — | — | — | — | — | |
| ObjectFusionModality=LC2026.05 | 71 | 73.3 | 89.4 | 59 | 71.8 | — | 40.5 | 90.7 | 78.1 | — | 87.7 | — | — | — | — | — | — | — | 53.2 | — | — | 80 | — | — | 63.1 | — | — | |
| ObjectFusionVenue=ICCV’23, Modality=LC2026.06 | 71 | 73.3 | 89.4 | 59 | 71.8 | 63.1 | 40.5 | 90.7 | 78.1 | — | 87.7 | 76.6 | — | — | — | — | — | — | 53.2 | — | — | — | — | — | — | — | — | |
| UniTR w/ LSSModality=Camera + LiDAR2023.08 | 70.9 | 74.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| UniTRModality=LiDAR + Camera2024.03 | 70.9 | 74.5 | 87.9 | 60.2 | 72.2 | 65.1 | 39.2 | 89.4 | 75.8 | — | 89.7 | — | — | — | — | — | — | — | 52.2 | — | — | 76.8 | — | — | — | — | — | |
| UniTRModality=L+C2024.07 | 70.9 | 74.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DepthFusionVenue=TMM’26, Modality=LC2026.06 | 70.9 | 74.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepInteraction-baseModality=L+C, Image Backbone=R50, LiDAR Backbone=VoxelNet, Test-time augmentation=false, Model ensemble=false2022.08 | 70.8 | 73.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepInteractionModality=LC, Single-model=true, Ensemble=false, Test-time augmentation=false2022.09 | 70.8 | 73.4 | 87.9 | 60.2 | 70.8 | 63.8 | 37.5 | 90.3 | 75.4 | — | 87 | 80.4 | — | — | — | — | — | — | 54.5 | — | — | — | — | — | — | — | — | |
| DeepInteractionDevice=V100 GPU2023.08 | 70.8 | 73.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 480 | — | — | — | — | — | — | |
| DeepInteractionModality=Camera + LiDAR2023.08 | 70.8 | 73.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepInteractionModality=C+L2023.03 | 70.8 | 73.4 | — | — | — | — | — | — | — | — | — | — | 0.257 | 0.24 | 0.325 | 0.245 | 0.128 | — | — | — | — | — | — | — | — | — | — | |
| DeepInteractionModality=LiDAR + Camera2024.03 | 70.8 | 73.4 | 87.9 | 60.2 | 70.8 | 63.8 | 37.5 | 91.7 | 75.4 | — | 87.2 | — | — | — | — | — | — | — | 54.5 | — | — | 80.4 | — | — | — | — | — | |
| DeepInteractionModality=L+C2024.07 | 70.8 | 73.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DeepInteractionModality=LC2024.11 | 70.8 | 73.4 | 87.9 | 60.2 | 70.8 | 63.8 | 37.5 | 91.7 | 75.4 | — | 87.2 | 80.4 | — | — | — | — | — | — | 54.5 | — | — | — | — | — | — | — | — | |
| DeepInteractionVenue=NeurIPS’22, Modality=LC2026.06 | 70.8 | 73.4 | 87.9 | 60.2 | 70.8 | 63.8 | 37.5 | 90.3 | 75.4 | — | 87 | 80.4 | — | — | — | — | — | — | 54.5 | — | — | — | — | — | — | — | — | |
| 3D Dual-Fusion (T)Modality=LiDAR-Camera based, Baseline=TransFusion2022.11 | 70.6 | 73.1 | 87.7 | 62 | — | — | 35.6 | 88.8 | 78 | — | — | 74.4 | — | — | — | — | — | — | 58 | — | — | — | — | — | — | — | — | |
| UniTRModality=Camera + LiDAR2023.08 | 70.5 | 74.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Multi-Modal Decouple and Recouple NetworkCondition=Clean2026.03 | 70.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CMTModality=C+L2023.03 | 70.4 | 73 | — | — | — | — | — | — | — | — | — | — | 0.299 | 0.241 | 0.323 | 0.24 | 0.112 | — | — | — | — | — | — | — | — | — | — | |
| CMTModality=LC2024.11 | 70.4 | 73 | 87.2 | 61.5 | 72.4 | 62.8 | 37.5 | 86.9 | 79.4 | — | 83.1 | 86.9 | — | — | — | — | — | — | 58.3 | — | — | — | — | — | — | — | — | |
| UniMamba2025.03 | 70.2 | 74 | 87.9 | 60.4 | 70.9 | 65.9 | 36.7 | 90.5 | 73.5 | 49.5 | 86.9 | 79.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LiDAR and Camera2022.11 | 70.2 | 72.9 | — | — | — | — | — | — | — | — | — | — | 0.239 | 0.329 | 0.26 | 0.134 | 0.153 | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionInput=C+L, Backbone=Swin-T, FPS=8.42022.09 | 70.2 | 72.9 | — | — | — | — | — | — | — | — | — | — | 0.261 | 0.239 | 0.329 | 0.26 | 0.134 | — | — | — | — | — | — | — | — | — | — | |
| BEVFusion [30]Modality=L+C, Image Backbone=Swin-Tiny, LiDAR Backbone=VoxelNet, Test-time augmentation=false, Model ensemble=false2022.08 | 70.2 | 72.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LiDAR-Camera based2022.11 | 70.2 | 72.9 | 88.6 | 60.1 | — | — | 39.3 | 89.2 | 74.1 | — | — | 80 | — | — | — | — | — | — | 51 | — | — | — | — | — | — | — | — | |
| BEVFusion [16]Modality=LC, Single-model=true, Ensemble=false, Test-time augmentation=false2022.09 | 70.2 | 72.9 | 88.6 | 60.1 | 69.8 | 63.8 | 39.3 | 89.2 | 74.1 | — | 86.5 | 80 | — | — | — | — | — | — | 51 | — | — | — | — | — | — | — | — | |
| BEVFusion (MIT)Modality=Camera + LiDAR2023.08 | 70.2 | 72.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LiDAR + Camera, citation_id=[42]2024.03 | 70.2 | 72.9 | 88.6 | 60.1 | 69.8 | 63.8 | 39.3 | 89.2 | 74.1 | — | 86.5 | — | — | — | — | — | — | — | 51 | — | — | 80 | — | — | — | — | — | |
| BEVFusionModality=L+C, Citation=[16]2024.07 | 70.2 | 72.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=C+L2022.05 | 70.2 | 72.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LC, citation=[26]2024.11 | 70.2 | 72.9 | 88.6 | 60.1 | 69.8 | 63.8 | 39.3 | 89.2 | 74.1 | — | 85.2 | 80 | — | — | — | — | — | — | 51 | — | — | — | — | — | — | — | — | |
| BEVFusionCondition=Clean2026.03 | 70.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LC, citation=[25]2026.05 | 70.2 | 72.9 | 88.6 | 60.1 | 69.8 | — | 39.3 | 89.2 | 74.1 | — | 86.5 | — | — | — | — | — | — | — | 51 | — | — | 80 | — | — | 63.8 | — | — | |
| BEVFusion-MITVenue=ICRA’23, Modality=LC2026.06 | 70.2 | 72.9 | 88.6 | 60.1 | 69.8 | 63.8 | 39.3 | 89.2 | 74.1 | — | 86.5 | 80 | — | — | — | — | — | — | 51 | — | — | — | — | — | — | — | — | |
| Focals Conv-FModality=L+C, Image Backbone=R50, LiDAR Backbone=VoxelNet-FocalsConv, Test-time augmentation=true, Model ensemble=false2022.08 | 70.1 | 73.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MetaBEVCondition=Clean2026.03 | 70 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FUTR3D + ALIGNQuery Init.=ALIGN, Test-time augmentation=false2025.12 | 69.9 | 72.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LinKTTA=true2023.03 | 69.8 | 73.4 | 87.3 | 60.2 | 69.8 | 65.9 | 34 | 88.2 | 78.8 | — | 83 | 76.8 | — | — | — | — | — | — | 54.3 | — | — | — | — | — | — | — | — | |
| LION-MambaPresent at=-2024.07 | 69.8 | 73.9 | 87.2 | 61.1 | 68.9 | 65 | 36.3 | 90 | 74 | — | 87.3 | 79.5 | — | — | — | — | — | — | 49.2 | — | — | — | — | — | — | — | — | |
| LION-Mamba2024.07 | 69.8 | 73.9 | 87.2 | 61.1 | 68.9 | — | 36.3 | 90 | — | — | 87.3 | — | — | — | — | — | — | — | 49.2 | — | — | 79.5 | — | — | 65 | 74 | — | |
| SCAFusionModality=Fusion2025.12 | 69.7 | 71.9 | 88.7 | 61.4 | 72.4 | 63.5 | 36.6 | 89.3 | 74.3 | — | 86.4 | 74.5 | — | — | — | — | — | — | 50.1 | — | — | — | — | — | — | — | — | |
| Fusion4CAModality=L+C2026.03 | 69.7 | 72.1 | 88.7 | 61.4 | 72.4 | — | 36.6 | 89.3 | — | 50.1 | 86.4 | — | — | — | — | — | — | — | — | — | — | 74.5 | — | — | 63.5 | 74.3 | — | |
| FUTR3DQuery Init.=Random, Test-time augmentation=false2025.12 | 69.4 | 72.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FUTR3DModality=L+C2026.03 | 69.4 | 72.1 | 86.3 | 61.5 | 71.9 | — | 26 | 82.6 | — | 63.3 | 70.1 | — | — | — | — | — | — | — | — | — | — | 64.4 | — | — | 42.1 | 73.6 | — | |
| FUTR3DCondition=Clean2026.03 | 69.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CMT-small + ALIGNQuery Init.=ALIGN, Test-time augmentation=false2025.12 | 69.3 | 72.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LC2022.05 | 69.2 | 71.8 | 88.1 | 60.9 | 69.3 | 62.1 | 34.4 | 89.2 | 72.2 | — | 85.2 | 78.2 | — | — | — | — | — | — | 52.2 | — | — | — | — | — | — | — | — | |
| BEVFusion [26]Modality=L+C, Image Backbone=Swin-Tiny, LiDAR Backbone=VoxelNet, Test-time augmentation=false, Model ensemble=false2022.08 | 69.2 | 71.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusion [13]Modality=LC, Single-model=true, Ensemble=false, Test-time augmentation=false2022.09 | 69.2 | 71.8 | 88.1 | 60.9 | 69.3 | 62.1 | 34.4 | 89.2 | 72.2 | — | 85.2 | 78.2 | — | — | — | — | — | — | 52.2 | — | — | — | — | — | — | — | — | |
| BEVFusionDevice=V100 GPU2023.08 | 69.2 | 71.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 1,610 | — | — | — | — | — | — | |
| BEVFusion (PKU)Modality=Camera + LiDAR2023.08 | 69.2 | 71.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFusionModality=LiDAR + Camera, citation_id=[37]2024.03 | 69.2 | 71.8 | 88.1 | 60.9 | 69.3 | 62.1 | 34.4 | 89.2 | 72.2 | — | 85.5 | — | — | — | — | — | — | — | 52.2 | — | — | 78.2 | — | — | — | — | — | |
| BEVFusion-PKUModality=Fusion2025.12 | 69.2 | 71.8 | 88.1 | 60.9 | 69.3 | 62.1 | 34.4 | 89.2 | 72.2 | — | 85.2 | 78.2 | — | — | — | — | — | — | 52.2 | — | — | — | — | — | — | — | — |