List your company

3D Instance Segmentation on ScanNet200

63.7mAP@0.5ESAM
1.71617.80833.949.992Apr 16, 2022Dec 15, 2022Aug 15, 2023Apr 14, 2024Dec 13, 2024Aug 13, 2025Apr 14, 2026

Evaluation results

MethodLinks
ESAMType=Online, 3D GT=Yes2026.0163.7——42.279.6————
3AMType=Online, 3D GT=No2026.0159.7——47.375.3————
SAM2ObjectType=Offline, 3D GT=Yes2026.0152.7——3470.3————
SegDINO3DInput Modality=P & I & D & C2026.0352.4———58.6———40.2
SAI3DType=Offline, 3D GT=Yes2026.0147.2——28.267.9————
SegVGGTInput Modality=I2026.0345.7———53.7———31.9
ODINInput Modality=I & D & C2026.0345.3———53.1———31.5
Open3DISType=Offline, 3D GT=Yes2026.0143.1——34.648.5————
OV-3DIS3D mask proposal=2D+3D, evaluation protocol=Top-k2026.0441.4———45.334.533.130.732.7
Relation3DInput Modality=P2026.0341.2———45.6———31.6
OneFormer3DInput Modality=P2026.0340.9———44.6———30.2
Mask3D3D mask proposal=3D, setting=closed-vocabulary2026.0439.8———21.717.9——26.9
SGIFormer-LInput Modality=P2026.0339.4———44.2———29.2
CAMFusion3D mask proposal=3D, evaluation protocol=Top-k2026.0439.3———46.231.829.122.227.6
MAFTInput Modality=P2026.0338.2———43.3———29.2
OV-3DIS3D mask proposal=3D, evaluation protocol=Top-k2026.0437.6———42.83325.328.129
QueryFormerInput Modality=P2026.0337.1———43.4———28.1
Mask3DInput Modality=P2026.0337———42.3———27.4
Mask3DSetting=Fully-sup2023.1236.2——26.941.439.817.921.7—
Mask3DMask supervision=supervised, Semantic supervision=supervised2024.1136.2——26.941.4————
SAM3DType=Online, 3D GT=Yes2026.0135.7——20.255.5————
CAMFusion3D mask proposal=3D, evaluation protocol=Top-12026.0435———4127.425.620.424.4
SAMPro3DType=Offline, 3D GT=Yes2026.0132.8——1856.1————
ISBNetSetting=Fully-sup2023.1232.7——24.537.638.612.520.5—
OV-3DIS3D mask proposal=2D+3D, evaluation protocol=Top-12026.0432.5———36.226.328.223.225.8
OV-3DIS3D mask proposal=3D, evaluation protocol=Top-12026.0431.8———36.427.223.122.324.2
Open-YOLO 3DInput Modality=P & I & D & C2026.0331.7———36.2———24.7
Open-YOLO3D3D mask proposal=3D, evaluation protocol=Top-k2026.0431.7———36.227.821.624.324.7
PE+TR+AvgPool3D mask proposal=3D, evaluation protocol=Top-k2026.0431———37.523.424.118.421.8
PE+TR+AvgPool3D mask proposal=3D, evaluation protocol=Top-12026.0431———37.523.424.118.421.8
Ours (Mask3D)Mask supervision=supervised, Semantic supervision=zero-shot, Model variant=Mask3D2024.1130.3——23.233.3————
Open3DISInputs=3D+2D, 3D Region Proposal Network=Superpoints + ISBNet + Grounded-SAM2025.1229.4———32.827.821.8——
Open3DIS (2D and 3D)Setting=Open-vocab, 3D Proposal=ISBNet [50]2023.1229.4——23.732.827.821.821.2—
Open3DISMask supervision=supervised, Semantic supervision=zero-shot2024.1129.4——23.732.8————
Open3DIS3D mask proposal=2D+3D, evaluation protocol=Top-k2026.0429.4———32.827.821.821.223.7
Open-YOLO3D3D mask proposal=3D, evaluation protocol=Top-12026.0428.3———31.725.618.52121.9
CLIP onlypre-training=CLIP embeddings2022.0427.9164.2127.58——————
Open3DIS (only 2D)Setting=Open-vocab, 3D Proposal=None2023.1226.1——18.231.418.919.216.5—
Language-Grounded 3D Pre-trainingbatch size=8, epochs=300, optimizer=momentum SGD, starting learning rate=0.022022.0426.0965.3227.72——————
OursMask supervision=zero-shot, Semantic supervision=zero-shot2024.1125.8——14.333.6————
CSCpre-training=CSC2022.0425.2463.1325.92——————
OVIR-3DSetting=Open-vocab, 3D Proposal=None2023.1224.9——1332.314.411.712.7—
Scratchpre-training=none2022.0424.4761.0425.37——————
Mask-ClusteringMask supervision=zero-shot, Semantic supervision=zero-shot2024.1123.3——1230.1————
Open3DIS (only 3D)Setting=Open-vocab, 3D Proposal=ISBNet [50]2023.1223.1——18.627.324.713.316.9—
Open3DIS (3D only)Mask supervision=supervised, Semantic supervision=zero-shot, Model variant=3D only2024.1123.1——18.627.3————
OpenMask3DSetting=Open-vocab, 3D Proposal=Mask3D [60]2023.1219.9——15.423.117.114.914.1—
OpenMask3D (Mask3D)Mask supervision=supervised, Semantic supervision=zero-shot, Model variant=Mask3D2024.1119.9——15.423.1————
OpenMask3DInput Modality=P & I & D & C2026.0319.9———23.1———15.4
SAI3DInputs=3D+2D, 3D Region Proposal Network=Superpoints + SAM2025.1218.8———24.112.116.2——
SAI3DMask supervision=zero-shot, Semantic supervision=zero-shot2024.1118.8——12.724.1————
OVIR-3DMask supervision=zero-shot, Semantic supervision=zero-shot2024.1118.7——9.325————
ChorusInputs=3D, 3D Region Proposal Network=Mask3D2025.1218———19.618.513——
Mosaic3DInputs=3D, 3D Region Proposal Network=Mask3D2025.1216———17.821.85.4——
OpenScene + Mask3DSetting=Open-vocab, 3D Proposal=Mask3D [60]2023.1215.2——11.717.813.49.911.6—
SAM3DMask supervision=zero-shot, Semantic supervision=zero-shot2024.1115.2——9.820.7————
SAM3DSetting=Open-vocab, 3D Proposal=None2023.1214.2——6.121.374.66.2—
OpenIns3DInputs=3D, 3D Region Proposal Network=Mask3D2025.1210.3———14.4164.2——
OpenIns3DSetting=Open-vocab, 3D Proposal=Mask3D [60]2023.1210.3——8.814.4164.26.5—
RegionPLCInputs=3D, 3D Region Proposal Network=Mask3D2025.128.6———9.715.61.7——
OpenScene + DBScanSetting=Open-vocab, 3D Proposal=None2023.127.8——2.818.62.72.63.1—
OpenScene-3DInputs=3D, 3D Region Proposal Network=Mask3D2025.126.2———7.210.60.7——
OneFormer3DInput Modality=I, Reconstruction=3D reconstruction with VGGT [52]2026.034.1———6.4———2.5
Any3DIS3D mask proposal=2D+3D, evaluation protocol=Top-k2026.04—————27.426.423.825.8
AvgPool w/ PE+TRevaluation protocol=Top-k2026.04—————57.75842.352.3
AvgPool w/ PE+TRevaluation protocol=Top-12026.04—————51.957.941.450
CAMFusionevaluation protocol=Top-k2026.04—————70.365.651.262.1
CAMFusionevaluation protocol=Top-12026.04—————60.961.346.756
OpenYolo3Devaluation protocol=Top-k2026.04—————43.338.536.839.6
OpenYolo3Devaluation protocol=Top-12026.04—————43.338.536.939.6
OV-3DISevaluation protocol=Top-k2026.04—————33.932.526.530.9
OV-3DISevaluation protocol=Top-12026.04—————30.33625.830.3
OV-3DIS w/ PE+TRevaluation protocol=Top-k2026.04—————51.150.538.946.6
OV-3DIS w/ PE+TRevaluation protocol=Top-12026.04—————43.754.538.745.1
Search3D3D mask proposal=3D, evaluation protocol=Top-12026.04—————26.321.421.223