List your company
ResearchFormerly WizWand

Object Detection on COCO (APbbox)

61.3AP (bbox)HIPIE
13.87626.18838.550.812Oct 20, 2022Apr 29, 2023Nov 7, 2023May 17, 2024Nov 24, 2024Jun 4, 2025Dec 13, 2025
Updated 7mo ago

Evaluation results

MethodLinks
HIPIEBackbone=ViT-H2023.0761.3
HIPIEBackbone=ViT-H2023.0761.3
APE (B)Backbone=ViT-L, Pre-training data=COCO, LVIS, O365, OID, VG, Evaluation Protocol=Supervised2024.0560.7
GLIPv2Backbone=Swin-H [32], Pre-training data=FourODS, COCO, GoldG, CC15M, SBU, Evaluation Protocol=Supervised2024.0560.6
MixPLDetector=DINO, Backbone=Swin-L2023.1260.2
AIMv2Evaluation protocol=Finetuning on grounding dataset mixture, Detection mode=Open-vocabulary2024.1160.2
DINOv2Evaluation protocol=Finetuning on grounding dataset mixture, Detection mode=Open-vocabulary2024.1160.1
Soft TeacherDetector=HTC++, Backbone=Swin-L2023.1259.9
DFN-CLIPEvaluation protocol=Finetuning on grounding dataset mixture, Detection mode=Open-vocabulary2024.1159.8
GLEE-ProBackbone=ViT-L [6], Pre-training data=GLEE-merged-10M (COCO, LVIS, etc.), Evaluation Protocol=Supervised2024.0559.6
LLMDetBackbone=Swin-L, Resolution=1000*1560, #Params=343M, FPS=2.1, Zero-shot=true2025.1259.2
OAI CLIPEvaluation protocol=Finetuning on grounding dataset mixture, Detection mode=Open-vocabulary2024.1159.1
SigLIPEvaluation protocol=Finetuning on grounding dataset mixture, Detection mode=Open-vocabulary2024.1158.8
APE (D)Backbone=ViT-L [6], Pre-training data=COCO, LVIS, O365, OID, VG, RefC, SA-1B, GoldG, PhraseCut, Evaluation Protocol=Supervised2024.0558.3
APE-D*training=partially trained on LVIS2025.1158.3
SAM 32025.1156.4
DINO-X2025.1156
LLMDetBackbone=Swin-T, Resolution=800*1333, #Params=172M, FPS=6, Zero-shot=true2025.1255.6
LLMDet-L2025.1155.6
Grounding DINOBackbone=Swin-L [19], Pre-training data=O365, OID, GoldG, Cap4M, COCO, RefC, Evaluation Protocol=Supervised2024.0555.5
TEC_MAEBackbone=ViT-B, Framework=Cascade MaskRCNN, Source Implementation=Li et al., 2022b2022.1054.6
WeDetect-LargeBackbone=ConvNext-L, Resolution=1280*1280, #Params=490M, FPS=6, Zero-shot=true2025.1254.5
Grounding DINO 1.5 ProBackbone=ViT-L [6], Pre-training data=Grounding-20M, Evaluation Protocol=Zero-shot2024.0554.3
MAEBackbone=ViT-B, Framework=Cascade MaskRCNN, Source Implementation=Li et al., 2022b2022.1054
HIPIEBackbone=RN502023.0753.9
HIPIEBackbone=RN502023.0753.9
OmDet-Turbo-BBackbone=ConvNeXt-B [20], Pre-training data=O365, GoldG, PhraseCut, Hake, HOI-A, Evaluation Protocol=Zero-shot2024.0553.4
VIC-MAEpre-train data=IN1K+K710+MiT, Backbone=ViT-B Mask R-CNN2023.0353.2
TEC_iBOTBackbone=ViT-B, Framework=Cascade MaskRCNN, Source Implementation=Zhou et al., 2022a2022.1052.7
VIC-MAEpre-train data=IN1K+K400, Backbone=ViT-B Mask R-CNN2023.0352.5
Grounding-DINOBackbone=Swin-L, Resolution=800*1333, #Params=343M, FPS=2.1, Zero-shot=true2025.1252.5
Grounding DINOBackbone=Swin-L, Pre-training data=O365, OID, GoldG, Evaluation Protocol=Zero-shot2024.0552.5
C-MAEpre-train data=IN1K, Backbone=ViT-B Mask R-CNN2023.0352.4
T-Rex2Backbone=Swin-L, Resolution=800*1333, Zero-shot=true2025.1252.2
T-Rex2 (text)Backbone=Swin-L, Pre-training data=O365, OID, GoldG, CC3M, SBU, LAION, Evaluation Protocol=Zero-shot2024.0552.2
WeDetect-BaseBackbone=ConvNext-B, Resolution=640*640, #Params=176M, FPS=35.1, Zero-shot=true2025.1252.1
iBOTBackbone=ViT-B, Framework=Cascade MaskRCNN, Source Implementation=Zhou et al., 2022a2022.1051.2
MM-GDINOBackbone=Swin-T, Resolution=800*1333, #Params=172M, FPS=6, Zero-shot=true2025.1250.4
MAEpre-train data=IN1K, Backbone=ViT-B Mask R-CNN2023.0350.3
OV-DINOBackbone=Swin-T, Resolution=800*1333, Zero-shot=true2025.1250.2
GLIPBackbone=Swin-L, Resolution=800*1333, #Params=430M, FPS=3.1, Zero-shot=true2025.1249.8
GLIPBackbone=Swin-L, Pre-training data=FourODs, GoldG, Cap24M, Evaluation Protocol=Zero-shot2024.0549.8
DINO-X EdgeBackbone=EfficientViT-L2, Resolution=640*640, FPS=19.8, Zero-shot=true2025.1248.7
DetCLIPv3Backbone=Swin-L, Resolution=800*1333, Zero-shot=true2025.1248.5
Grounding-DINOBackbone=Swin-T, Resolution=800*1333, #Params=172M, FPS=6, Zero-shot=true2025.1248.4
Grounding DINOBackbone=Swin-T, Pre-training data=O365, GoldG, Cap4M, Evaluation Protocol=Zero-shot2024.0548.4
DetCLIPv3Backbone=Swin-T, Resolution=800*1333, Zero-shot=true2025.1247.2
T-Rex2 (visual)Backbone=Swin-L, Pre-training data=O365, OID, HierText, CrowdHuman, SA-1B, Evaluation Protocol=Zero-shot2024.0546.5
ODISEBackbone=ViT-H+SD2023.0746.1
ODISEBackbone=ViT-H+SD2023.0746.1
GLIPBackbone=Swin-T, Resolution=800*1333, #Params=232M, FPS=5.4, Zero-shot=true2025.1246.1
OWLv2*training=partially trained on LVIS2025.1146.1
T-Rex2Backbone=Swin-T, Resolution=800*1333, Zero-shot=true2025.1245.8
gDino-T2025.1145.7
DetCLIPv3Backbone=Swin-L, Pre-training data=O365, V3Det, GoldG, GranuCap50M, Evaluation Protocol=Zero-shot2024.0545.1
YOLO-World-LBackbone=YOLOv8-L, Resolution=640*640, #Params=48M, FPS=54.6, Zero-shot=true2025.1244.9
WeDetect-TinyBackbone=ConvNext-T, Resolution=640*640, #Params=33M, FPS=62.5, Zero-shot=true2025.1244.9
DetCLIPv2Backbone=Swin-L, Pre-training data=O365, GoldG, CC15M, Evaluation Protocol=Zero-shot2024.0544.7
MQ-GLIPBackbone=Swin-L, Pre-training data=O365, Evaluation Protocol=Zero-shot2024.0543.4
OWL-ViTBackbone=ViT-L [5], Pre-training data=O365, OID, VG, LIT, Evaluation Protocol=Zero-shot2024.0542.2
OWL-STBackbone=CLIP L/14 [23], Pre-training data=WebLI2B, Evaluation Protocol=Zero-shot2024.0540.9
DetCLIPBackbone=Swin-L, Pre-training data=O365, GoldG, YFCC1M, Evaluation Protocol=Zero-shot2024.0538.6
OWLv22025.1138.2
YOLO-WorldBackbone=YOLOv8-L [10], Pre-training data=O365, GoldG, CC3M, Evaluation Protocol=Zero-shot2024.0535.4
OpenSeeDBackbone=Swin-L, Pre-training data=COCO, O365, Evaluation Protocol=Zero-shot2024.0523
DINOvBackbone=Swin-L, Pre-training data=COCO, SA-1B, Evaluation Protocol=Zero-shot2024.0515.7