3D Occupancy Prediction on Occ3D-nuScenes (val)
5,335mIoUTFusionOcc (T-SQ-IW-25600)
Evaluation Results
| Method | Links | ||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| TFusionOcc (T-SQ-IW-25600)Mask=true, Backbone=Res-50, Modality=C+L, primitive_type=T-Superquadric, IW=true, tokens=256002026.02 | 5,335 | — | — | — | — | — | 1,466 | 6,025 | 4,080 | 5,839 | 6,367 | 3,481 | 4,597 | 5,904 | 4,872 | 4,532 | 5,043 | 8,321 | 4,470 | 5,555 | 6,022 | 7,141 | 6,975 | — | — | — | — | — | — | — | — | — | — | — | |
| TFusionOcc (T-SQ-IW-12800)Mask=true, Backbone=Res-50, Modality=C+L, primitive_type=T-Superquadric, IW=true, tokens=128002026.02 | 5,297 | — | — | — | — | — | 1,490 | 6,059 | 4,093 | 5,648 | 6,368 | 3,525 | 4,596 | 5,837 | 4,862 | 4,454 | 4,984 | 8,342 | 4,385 | 5,509 | 5,967 | 6,986 | 6,951 | — | — | — | — | — | — | — | — | — | — | — | |
| EFFOccMask=true, Backbone=Res-50, Modality=C+L2026.02 | 5,282 | — | — | — | — | — | 1,209 | 5,967 | 3,339 | 6,176 | 6,498 | 3,546 | 4,601 | 5,709 | 4,104 | 4,787 | 5,459 | 8,276 | 4,395 | 5,637 | 6,023 | 7,112 | 6,960 | — | — | — | — | — | — | — | — | — | — | — | |
| TFusionOcc (T-SQ-12800)Mask=true, Backbone=Res-50, Modality=C+L, primitive_type=T-Superquadric, tokens=128002026.02 | 5,234 | — | — | — | — | — | 1,482 | 5,885 | 3,948 | 5,510 | 6,224 | 3,542 | 4,419 | 5,861 | 4,717 | 4,492 | 4,843 | 8,254 | 4,287 | 5,465 | 5,952 | 7,105 | 6,996 | — | — | — | — | — | — | — | — | — | — | — | |
| TFusionOcc (T-SQ-25600)Mask=true, Backbone=Res-50, Modality=C+L, primitive_type=T-Superquadric, tokens=256002026.02 | 5,170 | — | — | — | — | — | 1,489 | 5,919 | 3,764 | 5,396 | 6,240 | 3,388 | 4,474 | 5,841 | 4,717 | 4,108 | 4,723 | 8,246 | 4,325 | 5,480 | 5,904 | 7,034 | 6,839 | — | — | — | — | — | — | — | — | — | — | — | |
| SDGOccMask=true, Backbone=Res-50, Modality=C+L2026.02 | 5,166 | — | — | — | — | — | 1,320 | 5,780 | 2,430 | 6,030 | 6,430 | 3,620 | 3,940 | 5,240 | 3,580 | 5,090 | 5,370 | 8,460 | 4,750 | 5,800 | 6,160 | 7,070 | 6,770 | — | — | — | — | — | — | — | — | — | — | — | |
| TFusionOcc (T-P-25600)Mask=true, Backbone=Res-50, Modality=C+L, primitive_type=T-Primitive, tokens=256002026.02 | 5,151 | — | — | — | — | — | 1,475 | 5,810 | 3,773 | 5,121 | 6,203 | 3,555 | 4,482 | 5,518 | 4,370 | 4,433 | 4,924 | 8,271 | 4,259 | 5,483 | 5,927 | 7,021 | 6,941 | — | — | — | — | — | — | — | — | — | — | — | |
| RadOccMask=true, Backbone=Swin-B, Modality=C+L2026.02 | 4,938 | — | — | — | — | — | 1,090 | 5,820 | 2,500 | 5,790 | 6,290 | 3,400 | 3,350 | 5,010 | 3,210 | 4,890 | 5,210 | 8,290 | 4,270 | 5,530 | 5,830 | 6,860 | 6,600 | — | — | — | — | — | — | — | — | — | — | — | |
| OccFusionMask=true, Backbone=R101-DCN, Modality=C+L2026.02 | 4,874 | — | — | — | — | — | 1,240 | 5,180 | 3,300 | 5,460 | 5,770 | 3,400 | 4,300 | 4,840 | 3,550 | 4,120 | 4,860 | 8,300 | 4,470 | 5,710 | 6,000 | 6,250 | 6,130 | — | — | — | — | — | — | — | — | — | — | — | |
| RM2OccMask=true, Backbone=R101-DCN, Modality=C+L2026.02 | 4,782 | — | — | — | — | — | 1,334 | 5,453 | 3,381 | 5,830 | 5,997 | 3,445 | 3,489 | 2,943 | 3,153 | 3,991 | 4,244 | 9,601 | 4,976 | 6,128 | 6,762 | 5,371 | 5,196 | — | — | — | — | — | — | — | — | — | — | — | |
| OccFusionMask=true, Backbone=R101-DCN, Modality=C+L+R2026.02 | 4,667 | — | — | — | — | — | 1,240 | 5,030 | 3,150 | 5,760 | 5,880 | 3,400 | 4,100 | 4,720 | 2,970 | 4,200 | 4,800 | 7,840 | 3,570 | 4,730 | 5,270 | 6,350 | 6,330 | — | — | — | — | — | — | — | — | — | — | — | |
| GaussianFormer3DMask=true, Backbone=R101-DCN, Modality=C+L2026.02 | 4,640 | — | — | — | — | — | 980 | 5,000 | 3,130 | 5,400 | 5,940 | 2,810 | 3,620 | 4,620 | 2,670 | 4,020 | 4,970 | 7,910 | 3,730 | 4,900 | 5,500 | 6,910 | 6,760 | — | — | — | — | — | — | — | — | — | — | — | |
| Inverse++Mask=false, Backbone=Res-101, Modality=C2026.02 | 3,104 | — | — | — | — | — | 956 | 4,191 | 2,353 | 4,238 | 4,635 | 1,861 | 2,803 | 2,661 | 2,477 | 2,593 | 3,481 | 6,010 | 3,323 | 3,762 | 3,483 | 1,920 | 2,026 | — | — | — | — | — | — | — | — | — | — | — | |
| CTF-OccMask=false, Backbone=Res-101, Modality=C2026.02 | 2,853 | — | — | — | — | — | 809 | 3,933 | 2,056 | 3,829 | 4,224 | 1,693 | 2,452 | 2,272 | 2,105 | 2,298 | 3,111 | 5,333 | 3,384 | 3,798 | 3,323 | 2,079 | 1,800 | — | — | — | — | — | — | — | — | — | — | — | |
| TPVFormerMask=false, Backbone=Res-101, Modality=C2026.02 | 2,834 | — | — | — | — | — | 667 | 3,920 | 1,424 | 4,154 | 4,698 | 1,921 | 2,264 | 1,787 | 1,454 | 3,020 | 3,551 | 5,618 | 3,365 | 3,569 | 3,161 | 1,997 | 1,612 | — | — | — | — | — | — | — | — | — | — | — | |
| RenderOccMask=false, Backbone=Swin-B, Modality=C2026.02 | 2,611 | — | — | — | — | — | 484 | 3,172 | 1,072 | 2,767 | 2,645 | 1,387 | 1,820 | 1,767 | 1,784 | 2,119 | 2,325 | 6,320 | 3,642 | 4,621 | 4,426 | 1,958 | 2,072 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVStereoMask=false, Backbone=Res-101, Modality=C2026.02 | 2,451 | — | — | — | — | — | 573 | 3,841 | 788 | 3,870 | 4,120 | 1,756 | 1,733 | 1,469 | 1,031 | 1,684 | 2,962 | 5,408 | 2,892 | 3,268 | 2,654 | 1,874 | 1,749 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFormerMask=false, Backbone=Res-101, Modality=C2026.02 | 2,367 | — | — | — | — | — | 503 | 3,879 | 998 | 3,441 | 4,109 | 1,324 | 1,650 | 1,815 | 1,783 | 1,866 | 2,770 | 4,895 | 2,773 | 2,908 | 2,538 | 1,541 | 1,446 | — | — | — | — | — | — | — | — | — | — | — | |
| OccFormerMask=false, Backbone=Res-101, Modality=C2026.02 | 2,193 | — | — | — | — | — | 594 | 3,029 | 1,232 | 3,440 | 3,917 | 1,444 | 1,645 | 1,722 | 927 | 1,390 | 2,636 | 5,099 | 3,096 | 3,466 | 2,273 | 676 | 697 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDetMask=false, Backbone=Res-101, Modality=C2026.02 | 1,173 | — | — | — | — | — | 209 | 1,529 | 0 | 418 | 1,297 | 135 | 0 | 43 | 13 | 659 | 666 | 5,272 | 1,904 | 2,645 | 2,178 | 1,451 | 1,526 | — | — | — | — | — | — | — | — | — | — | — | |
| MonoSceneMask=false, Backbone=Res-101, Modality=C2026.02 | 606 | — | — | — | — | — | 175 | 723 | 426 | 493 | 938 | 567 | 398 | 301 | 590 | 445 | 717 | 1,491 | 632 | 792 | 743 | 101 | 765 | — | — | — | — | — | — | — | — | — | — | — | |
| CascadeOcc2026.06 | 69.34 | — | — | — | — | — | 54.82 | 77.9 | 76.41 | 72.27 | 72.96 | 51.07 | 79.21 | 79.27 | 75.34 | 61.25 | 68.48 | 84.16 | 82 | 72.37 | 69.64 | 55.35 | 46.32 | 64.12 | — | — | — | — | — | — | — | — | — | — | |
| OccWorld2026.06 | 64.74 | — | — | — | — | — | 46.34 | 71.84 | 69.96 | 67.59 | 69.01 | 45.14 | 73.5 | 74.77 | 68.57 | 54.65 | 65.27 | 82.74 | 78.18 | 69.81 | 66.53 | 52.95 | 43.77 | 61.88 | — | — | — | — | — | — | — | — | — | — | |
| Gau-OccModality=C+L2026.03 | 55.1 | — | — | — | — | — | 14.8 | 61.2 | 40.7 | 65.3 | 68.1 | 37.5 | 49.6 | 58.4 | 43.7 | 51.3 | 55.4 | 84.5 | 48.3 | 58.5 | 60.3 | 71.5 | 68 | — | — | — | — | — | — | — | — | — | — | — | |
| DAOccModality=Camera + LiDAR, Image Size=256 x 704, 2D Backbone=R502024.09 | 54.33 | — | 7.8 | — | — | — | 13 | 60.7 | 39.8 | 64 | 66.5 | 36.3 | 49 | 60.1 | 44.3 | 50.7 | 55.9 | 82.9 | 44.6 | 56.8 | 60.6 | 70.1 | 68.3 | — | — | — | — | — | — | — | — | — | — | — | |
| DAOccModality=C+L2026.03 | 54.3 | — | — | — | — | — | 13 | 60.7 | 39.8 | 64 | 66.5 | 36.3 | 49 | 60.1 | 44.3 | 50.7 | 55.9 | 82.9 | 44.6 | 56.8 | 60.6 | 70.1 | 68.3 | — | — | — | — | — | — | — | — | — | — | — | |
| SurroundSDFBackbone=ConvNext-B, Image Size=640 x 1600, Test Time Augmentation=false2024.03 | 53.01 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCC (Version K)backbone=InternImage-H, parameters=1200.0M2023.07 | 52.79 | — | — | — | — | — | 28.28 | 56.7 | 44.35 | 51.37 | 61.81 | 35.12 | 47.38 | 41.56 | 39.88 | 57.96 | 48.39 | 86.66 | 56.97 | 64.66 | 61.23 | 62.78 | 52.35 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCCBackbone=InternImage-H, Image Size=960 x 1769, Test Time Augmentation=true2024.03 | 52.79 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SDGOccModality=C+L2026.03 | 51.7 | — | — | — | — | — | 13.2 | 57.8 | 34.8 | 60.3 | 64.3 | 36.2 | 39.4 | 52.4 | 35.8 | 50.9 | 53.7 | 84.7 | 47.5 | 58 | 61 | 70.7 | 67.7 | — | — | — | — | — | — | — | — | — | — | — | |
| SDGOCCModality=Camera + LiDAR, 2D Backbone=R502024.09 | 51.66 | — | 7.5 | — | — | — | 13.2 | 57.8 | 24.3 | 60.3 | 64.3 | 36.2 | 39.4 | 52.4 | 35.8 | 50.9 | 53.7 | 84.6 | 47.5 | 58 | 61.6 | 70.7 | 67.7 | — | — | — | — | — | — | — | — | — | — | — | |
| UniOCCBackbone=ConvNext-B, Image Size=640 x 1600, Test Time Augmentation=false2024.03 | 51.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ALOccModality=Camera, Image Size=512 x 1408, 2D Backbone=Swin-B, Temporal Fusion=16 frames2024.09 | 50.6 | — | 1.5 | — | — | — | 17 | 58.3 | 39.7 | 56.6 | 63.2 | 33.2 | 41.3 | 40.3 | 40.8 | 43.8 | 51 | 87 | 52.7 | 62 | 65.2 | 57.7 | 50.9 | — | — | — | — | — | — | — | — | — | — | — | |
| ALOCCModality=C2026.03 | 50.6 | — | — | — | — | — | 16.9 | 55 | 37.5 | 56.9 | 62.9 | 32.4 | 39 | 34.8 | 34.8 | 41 | 50.1 | 87 | 53.7 | 62.7 | 65.2 | 57 | 50.9 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCC (Version J)backbone=ViT-L, adapter=ViT-Adapter, parameters=428.8M2023.07 | 50.47 | — | — | — | — | — | 16.74 | 55.33 | 39.77 | 58.94 | 61.79 | 32.04 | 42.63 | 40.51 | 39.06 | 43.72 | 51.33 | 87.34 | 53.77 | 62.63 | 66.06 | 56.63 | 49.74 | — | — | — | — | — | — | — | — | — | — | — | |
| RadOccModality=C2026.03 | 49.4 | — | — | — | — | — | 15.6 | 55.9 | 35 | 52 | 62.4 | 33.1 | 38.1 | 32.6 | 33.2 | 39.5 | 48 | 87.2 | 49.8 | 57.6 | 62.5 | 53.7 | 47.7 | — | — | — | — | — | — | — | — | — | — | — | |
| RadOccModality=Camera + LiDAR, Image Size=512 x 1408, 2D Backbone=Swin-B2024.09 | 49.38 | — | — | — | — | — | 10.9 | 58.2 | 25 | 57.9 | 62.9 | 34 | 33.5 | 50.1 | 32.1 | 48.9 | 52.1 | 82.9 | 42.7 | 55.3 | 58.3 | 68.6 | 66 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCCModality=Camera, Image Size=960 x 1760, 2D Backbone=VoVNet-99, Temporal Fusion=16 frames2024.09 | 48.9 | — | — | — | — | — | 14.3 | 57 | 38.3 | 57.7 | 62.1 | 34.4 | 39.4 | 38.8 | 39.4 | 42.9 | 50 | 86 | 50.2 | 60.1 | 62.5 | 52.4 | 45.7 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCC (Version I)backbone=VoVNet-99, input size=960x1760, voxel resolution=0.4m, parameters=130.8M2023.07 | 48.9 | — | — | — | — | — | 14.26 | 57.02 | 38.34 | 57.69 | 62.12 | 34.35 | 39.43 | 38.82 | 39.42 | 42.91 | 50.02 | 86.04 | 50.24 | 60.06 | 62.54 | 52.36 | 45.68 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCCModality=C2026.03 | 48.9 | — | — | — | — | — | 14.3 | 57 | 38.3 | 57.7 | 62.1 | 34.4 | 39.4 | 38.8 | 39.4 | 42.9 | 50 | 86 | 50.2 | 60.1 | 62.5 | 52.4 | 45.7 | — | — | — | — | — | — | — | — | — | — | — | |
| OccFusionModality=Camera + LiDAR, Image Size=900 x 1600, 2D Backbone=R1012024.09 | 48.74 | — | — | — | — | — | 12.4 | 51.8 | 33 | 54.6 | 57.7 | 34 | 43 | 48.4 | 35.5 | 41.2 | 48.6 | 83 | 44.7 | 57.1 | 60 | 62.5 | 61.3 | — | — | — | — | — | — | — | — | — | — | — | |
| OccFusionModality=C+L+R2026.03 | 48.7 | — | — | — | — | — | 14.2 | 51.8 | 33.3 | 54 | 57.7 | 34 | 43 | 48.4 | 35.5 | 41.2 | 48.6 | 83.4 | 57.1 | 60 | 62.5 | 61.3 | 44.9 | — | — | — | — | — | — | — | — | — | — | — | |
| DAOccModality (train)=Camera+LiDAR, Modality (val)=Camera+LiDAR, Image Size=256 × 704, 2D Backbone=R502024.09 | 48.3 | 48.4 | 7.8 | 44.5 | 48.9 | 51.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DAOccModality (train)=C+L, Modality (val)=C+L, Image Size=256 × 704, 2D Backbone=R502024.09 | 48.3 | 48.4 | 7.8 | 44.5 | 48.9 | 51.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OccFusionModality=Camera + LiDAR + Radar, Image Size=900 x 1600, 2D Backbone=R101-DCN2024.09 | 46.67 | — | 1 | — | — | — | 12.4 | 50.3 | 31.5 | 57.6 | 58.8 | 34 | 41 | 47.2 | 29.7 | 42 | 48 | 78.4 | 35.7 | 47.3 | 52.7 | 63.5 | 63.3 | — | — | — | — | — | — | — | — | — | — | — | |
| LMPOcc-LHistory Frame=1, Resolution=512 × 1408, Backbone=SwinB2025.04 | 46.61 | — | — | — | — | — | 13.68 | 53.88 | 31.65 | 53.19 | 58.53 | 30.68 | 34.7 | 34.86 | 34.6 | 39.45 | 47.17 | 85.08 | 47.85 | 58.11 | 61.56 | 57.36 | 49.97 | — | — | — | — | — | — | — | — | — | — | — | |
| COTRModality=Camera, Image Size=512 x 1408, 2D Backbone=Swin-B, Temporal Fusion=8 frames2024.09 | 46.21 | — | 0.3 | — | — | — | 14.9 | 53.3 | 35.2 | 50.8 | 57.3 | 35.4 | 34.1 | 33.5 | 37.1 | 39 | 45 | 84.5 | 48.7 | 57.6 | 61.1 | 51.6 | 46.7 | — | — | — | — | — | — | — | — | — | — | — | |
| COTRHistory Frame=8, Resolution=512 × 1408, Backbone=SwinB2025.04 | 46.2 | — | — | — | — | — | 14.85 | 53.25 | 35.19 | 50.83 | 57.25 | 35.36 | 34.06 | 33.54 | 37.14 | 38.99 | 44.97 | 84.46 | 48.73 | 57.6 | 61.08 | 51.61 | 46.72 | — | — | — | — | — | — | — | — | — | — | — | |
| COTRModality=C2026.03 | 46.2 | — | — | — | — | — | 14.9 | 53.3 | 35.2 | 50.8 | 57.4 | 34.1 | 33.5 | 31.9 | 33.1 | 39.1 | 45 | 84.5 | 48.3 | 57.6 | 61.1 | 51.6 | 46.7 | — | — | — | — | — | — | — | — | — | — | — | |
| SurroundSDFBackbone=ResNet101, Image Size=640 x 1600, Test Time Augmentation=false2024.03 | 46 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LightOcc-LHistory Frame=1, Resolution=512 × 1408, Backbone=SwinB2025.04 | 46 | — | — | — | — | — | 14.5 | 52.27 | 34.45 | 53.79 | 57.33 | 31.8 | 35.83 | 33.6 | 36.09 | 39.89 | 46.09 | 84.23 | 48.1 | 57.14 | 60.02 | 51.7 | 45.23 | — | — | — | — | — | — | — | — | — | — | — | |
| DHD-LHistory Frame=1, Resolution=512 × 1408, Backbone=SwinB2025.04 | 45.53 | — | — | — | — | — | 14.08 | 53.12 | 32.39 | 52.44 | 57.35 | 30.83 | 35.24 | 33.01 | 33.43 | 37.9 | 45.34 | 84.61 | 47.96 | 57.39 | 60.32 | 52.27 | 46.24 | — | — | — | — | — | — | — | — | — | — | — | |
| STCOccModality=Camera, Image Size=512 x 1408, 2D Backbone=R50, Temporal Fusion=16 frames2024.09 | 45 | — | 1.4 | — | — | — | 15.2 | 52.3 | 32.2 | 50.5 | 56.5 | 31.7 | 33.9 | 33.4 | 33.8 | 38.9 | 44.9 | 83.9 | 47.4 | 57.1 | 60.1 | 50.6 | 42.7 | — | — | — | — | — | — | — | — | — | — | — | |
| STCOccModality=C2026.03 | 45 | — | — | — | — | — | 15.2 | 52.3 | 32.2 | 50.5 | 56.3 | 31.7 | 34.3 | 31.3 | 33.8 | 39.3 | 44.9 | 84.3 | 47.1 | 57.1 | 61 | 52.4 | 46.3 | — | — | — | — | — | — | — | — | — | — | — | |
| GEOccHistory Frame=8, Resolution=512 × 1408, Backbone=SwinB2025.04 | 44.67 | — | — | — | — | — | 14.02 | 51.4 | 33.08 | 52.08 | 56.72 | 30.04 | 33.54 | 32.34 | 35.83 | 39.34 | 44.18 | 83.49 | 46.77 | 55.72 | 58.94 | 48.85 | 43 | — | — | — | — | — | — | — | — | — | — | — | |
| COTR(+Dr.Occ)Backbone=R50, Input Size=704×256, Camera mask supervision=true2026.03 | 44.1 | — | — | — | — | — | 14 | 53 | 32.6 | 45.8 | 54.9 | 29.7 | 31.5 | 30.9 | 35.7 | 38 | 42.5 | 83.8 | 44.9 | 56.1 | 58.8 | 51.1 | 45.6 | — | — | — | — | — | — | — | — | — | — | — | |
| OctreeOccImage Backbone=ResNet-101, Training with camera mask=true2023.12 | 44.02 | — | — | — | — | — | — | 51.7 | 29.93 | 53.52 | 56.77 | 30.83 | 33.17 | 30.65 | 29.99 | 37.76 | 43.87 | 83.17 | 44.52 | 55.45 | 58.86 | 49.52 | 46.33 | — | — | — | — | — | — | — | — | — | — | — | |
| OctreeOccHistory Frame=4, Resolution=512 × 1408, Backbone=R1012025.04 | 44.02 | — | — | — | — | — | 11.96 | 51.7 | 29.93 | 53.52 | 56.77 | 30.83 | 33.17 | 30.65 | 29.99 | 37.76 | 43.87 | 83.17 | 44.52 | 55.45 | 58.86 | 49.52 | 46.33 | — | — | — | — | — | — | — | — | — | — | — | |
| FlashOccModality=Camera, Image Size=512 x 1408, 2D Backbone=Swin-B, Temporal Fusion=2 frames, EMA=true2024.09 | 43.52 | — | 1.2 | — | — | — | 13.4 | 51.1 | 27.7 | 51.6 | 56.2 | 27.3 | 30 | 29.9 | 29.8 | 37.8 | 43.5 | 83.8 | 46.6 | 56.2 | 59.6 | 50.8 | 44.7 | — | — | — | — | — | — | — | — | — | — | — | |
| FlashOcc (M3)History Frame=1, Resolution=512 × 1408, Backbone=SwinB2025.04 | 43.52 | — | — | — | — | — | 13.42 | 51.07 | 27.68 | 51.57 | 56.22 | 27.27 | 29.98 | 29.93 | 29.8 | 37.77 | 43.52 | 83.81 | 46.55 | 56.15 | 59.56 | 50.84 | 44.67 | — | — | — | — | — | — | — | — | — | — | — | |
| FlashOccModality=C2026.03 | 43.5 | — | — | — | — | — | 13.4 | 51.8 | 32.1 | 54.1 | 58.3 | 29.9 | 32.8 | 31.9 | 29.9 | 38.7 | 46.2 | 85 | 47.4 | 56.2 | 59.6 | 50 | 43.7 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCCImage Backbone=ResNet-101, Training with camera mask=false2023.12 | 43.41 | — | — | — | — | — | — | 50.23 | 32.31 | 48.55 | 52.89 | 31.2 | 31.25 | 30.78 | 32.33 | 37.06 | 40.22 | 83.34 | 49.27 | 57.13 | 59.88 | 47.67 | 41.76 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OccHistory Frame=4, Resolution=512 × 1408, Backbone=R1012025.04 | 43.41 | — | — | — | — | — | 12.1 | 50.23 | 32.31 | 48.55 | 52.89 | 31.2 | 31.25 | 30.78 | 32.33 | 37.06 | 40.22 | 83.34 | 49.27 | 57.13 | 59.88 | 47.67 | 41.76 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDet4D(+Dr.Occ)Backbone=R50, Input Size=704×256, Camera mask supervision=true2026.03 | 43.4 | — | — | — | — | — | 14.9 | 52.8 | 30.7 | 45.6 | 55.1 | 30.3 | 30.8 | 31 | 35.9 | 39.1 | 42.5 | 81.1 | 43.4 | 54 | 57.6 | 49 | 44.5 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFormerBackbone=InterImage-XL, Image Size=640 x 1600, Test Time Augmentation=false2024.03 | 43.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDetBackbone=Swin-B, Image Size=640 x 1600, Test Time Augmentation=false2024.03 | 43.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| COTRBackbone=R50, Input Size=704×256, Camera mask supervision=true, Re-trained=true2026.03 | 43.1 | — | — | — | — | — | 12.8 | 51.6 | 30.3 | 45.4 | 54.9 | 30.4 | 30.4 | 30 | 33.9 | 36.9 | 42.1 | 83.4 | 44.4 | 55.3 | 58 | 49.6 | 44 | — | — | — | — | — | — | — | — | — | — | — | |
| SurroundSDFBackbone=ResNet50, Image Size=256 x 704, Test Time Augmentation=false2024.03 | 42.37 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PanoOccModality=Camera, Image Size=864 x 1600, 2D Backbone=R101-DCN, Temporal Fusion=4 frames2024.09 | 42.13 | — | 3 | — | — | — | 11.7 | 50.5 | 29.6 | 49.4 | 55.5 | 23.3 | 33.3 | 30.6 | 31 | 34.4 | 42.6 | 83.3 | 44.2 | 54.4 | 56 | 45.9 | 40.4 | — | — | — | — | — | — | — | — | — | — | — | |
| PanoOccImage Backbone=ResNet-101, Training with camera mask=true2023.12 | 42.13 | — | — | — | — | — | — | 50.48 | 29.64 | 49.44 | 55.52 | 23.29 | 33.26 | 30.55 | 30.99 | 34.43 | 42.57 | 83.31 | 44.23 | 54.4 | 56.04 | 45.94 | 40.4 | — | — | — | — | — | — | — | — | — | — | — | |
| PanoOccHistory Frame=3, Resolution=512 × 1408, Backbone=R1012025.04 | 42.13 | — | — | — | — | — | 11.67 | 50.48 | 29.64 | 49.44 | 55.52 | 23.29 | 33.26 | 30.55 | 30.99 | 34.43 | 42.57 | 83.31 | 44.23 | 54.4 | 56.04 | 45.94 | 40.4 | — | — | — | — | — | — | — | — | — | — | — | |
| PanoOccModality=C2026.03 | 42.1 | — | — | — | — | — | 11.7 | 50.5 | 29.4 | 49.4 | 55.5 | 24.3 | 33.3 | 30.6 | 31 | 34.4 | 42.6 | 83.3 | 44.2 | 54.4 | 56 | 45.9 | 40.4 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCC (Version H)backbone=ResNet-50, test-time augmentation=true, parameters=67.8M2023.07 | 42.06 | — | — | — | — | — | 14.3 | 49.71 | 30 | 46.62 | 51.54 | 29.3 | 29.13 | 29.35 | 30.48 | 34.97 | 39.36 | 83.07 | 47.16 | 55.62 | 59.88 | 44.89 | 39.58 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCCBackbone=ResNet50, Image Size=256 x 704, Test Time Augmentation=true2024.03 | 42.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDetOccModality=Camera, Image Size=512 x 1408, 2D Backbone=Swin-B, Temporal Fusion=2 frames2024.09 | 42.02 | — | 1.2 | — | — | — | 12.2 | 49.6 | 25.1 | 52 | 54.5 | 27.9 | 28 | 28.9 | 27.2 | 36.4 | 42.2 | 82.3 | 43.3 | 54.6 | 57.9 | 48.6 | 43.6 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDet4DImage Backbone=Swin-B, Training with camera mask=true2023.12 | 42.02 | — | — | — | — | — | — | 49.63 | 25.1 | 52.02 | 54.46 | 27.87 | 27.99 | 28.94 | 27.23 | 36.43 | 42.22 | 82.31 | 43.29 | 54.46 | 57.9 | 48.61 | 43.55 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDetOccModality=C2026.03 | 42 | — | — | — | — | — | 12.2 | 49.6 | 25.1 | 52 | 54.5 | 27.9 | 28 | 28.9 | 27.2 | 36.4 | 42.2 | 82.3 | 43.3 | 54.6 | 57.9 | 48.6 | 43.6 | — | — | — | — | — | — | — | — | — | — | — | |
| ViewFormerBackbone=R50, Input Size=704×256, Camera mask supervision=true2026.03 | 41.9 | — | — | — | — | — | 12.9 | 50.1 | 27.9 | 44.6 | 52.9 | 22.4 | 29.6 | 28 | 29.2 | 35.2 | 39.4 | 84.7 | 49.4 | 57.4 | 59.7 | 47.4 | 40.6 | — | — | — | — | — | — | — | — | — | — | — | |
| OSPBackbone=R101, Input Size=1600×900, Camera mask supervision=true2026.03 | 41.2 | — | — | — | — | — | 10.9 | 49 | 27.7 | 50.2 | 55.9 | 22.9 | 31 | 30.9 | 30.3 | 35.6 | 31.2 | 82.1 | 42.6 | 51.9 | 55.1 | 44.8 | 38.2 | — | — | — | — | — | — | — | — | — | — | — | |
| SurroundOccBackbone=InterImage-B, Image Size=640 x 1600, Test Time Augmentation=false2024.03 | 40.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCC (Version G)backbone=ResNet-50, dice loss=true, 3d transformation alignment=true2023.07 | 40.69 | — | — | — | — | — | 14.41 | 45.77 | 29.19 | 45.29 | 50.53 | 27.86 | 29.01 | 28.15 | 28.61 | — | 37.86 | 81.76 | 45.52 | 53.99 | 58.69 | 43.49 | 38.75 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCCBackbone=ResNet50, Image Size=256 x 704, Test Time Augmentation=false2024.03 | 40.69 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFormerBackbone=ResNet101, Image Size=900 x 1600, Test Time Augmentation=false2024.03 | 40.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| VoxFormerBackbone=ResNet101, Image Size=900 x 1600, Test Time Augmentation=false2024.03 | 40.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MiLOBackbone=ResNet50, Image Size=256 x 704, Test Time Augmentation=false2024.03 | 40.49 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LMPOcc-SHistory Frame=✗, Resolution=256 × 704, Backbone=R502025.04 | 40.38 | — | — | — | — | — | 10.97 | 48.87 | 23.66 | 43.31 | 51.27 | 23.61 | 23.79 | 27.49 | 26.28 | 36.26 | 37.95 | 81.97 | 42.06 | 52.09 | 58.43 | 52.96 | 45.45 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCC (Version F)backbone=ResNet-50, joint depth/semantic pre-training=true2023.07 | 39.89 | — | — | — | — | — | 13.66 | 45.88 | 28.26 | 44.91 | 49.78 | 26.21 | 28.84 | 28.27 | 27.89 | 32.75 | 37.56 | 81.2 | 43.46 | 52.13 | 56.35 | 42.79 | 38.1 | — | — | — | — | — | — | — | — | — | — | — | |
| UniOCCBackbone=ResNet50, Image Size=256 x 704, Test Time Augmentation=false2024.03 | 39.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDetOcc(8F)Modality (train)=Camera+LiDAR, Modality (val)=Camera, Image Size=384 × 704, 2D Backbone=R50, Temporal Frames=8F2024.09 | 39.3 | 32.6 | — | 26.6 | 33.1 | 38.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVDetOcc(8F)Modality (train)=C+L, Modality (val)=C, Image Size=384 × 704, 2D Backbone=R502024.09 | 39.3 | 32.6 | — | 26.6 | 33.1 | 38.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FastOccHistory Frame=16, Resolution=640 × 1600, Backbone=R1012025.04 | 39.21 | — | — | — | — | — | 12.06 | 43.53 | 28.04 | 44.8 | 52.16 | 22.96 | 29.14 | 29.68 | 26.98 | 30.81 | 38.44 | 82.04 | 41.93 | 51.92 | 53.71 | 41.04 | 35.49 | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFormer(4F)Modality (train)=Camera, Image Size=900 × 1600, 2D Backbone=R101, Temporal Frames=4F2024.09 | 39.2 | 32.4 | — | 26.1 | 32.9 | 38 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVFormer(4F)Modality (train)=C, Image Size=900 × 1600, 2D Backbone=R1012024.09 | 39.2 | 32.4 | — | 26.1 | 32.9 | 38 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OCC (Version E)backbone=ResNet-50, temporal frames=162023.07 | 39.11 | — | — | — | — | — | 13.57 | 44.74 | 27.01 | 45.41 | 49.1 | 25.15 | 26.33 | 27.86 | 27.79 | 32.28 | 36.75 | 80.07 | 42.76 | 51.18 | 55.13 | 42.19 | 37.53 | — | — | — | — | — | — | — | — | — | — | — | |
| FB-Occ(16F)Modality (train)=Camera+LiDAR, Modality (val)=Camera, Image Size=256 × 704, 2D Backbone=R50, Temporal Frames=16F2024.09 | 39.1 | 33.5 | — | 26.7 | 34.1 | 39.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FB-Occ(16F)Modality (train)=C+L, Modality (val)=C, Image Size=256 × 704, 2D Backbone=R502024.09 | 39.1 | 33.5 | — | 26.7 | 34.1 | 39.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BEVStereoBackbone=ResNet50, Image Size=384 x 704, Test Time Augmentation=false2024.03 | 39.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FB-OccBackbone=R50, Input Size=704×256, Camera mask supervision=true2026.03 | 39.1 | — | — | — | — | — | 13.6 | 44.7 | 27 | 45.4 | 49.1 | 25.2 | 26.3 | 27.9 | 27.8 | 32.3 | 36.8 | 80.1 | 42.8 | 51.2 | 55.1 | 42.2 | 37.5 | — | — | — | — | — | — | — | — | — | — | — | |
| ALOcc(16F)Modality (train)=Camera+LiDAR, Modality (val)=Camera, Image Size=256 × 704, 2D Backbone=R50, Temporal Frames=16F2024.09 | 38 | 43.7 | 6 | 37.8 | 44.7 | 48.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ALOcc(16F)Modality (train)=C+L, Modality (val)=C, Image Size=256 × 704, 2D Backbone=R502024.09 | 38 | 43.7 | 6 | 37.8 | 44.7 | 48.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |