List your company
ResearchFormerly WizWand

Image Classification on Pascal VOC 2007 (test)

92.6AccuracySEER
63.677671.186378.69586.2037Mar 2, 2021Jun 16, 2021Oct 1, 2021Jan 15, 2022May 2, 2022Aug 16, 2022Dec 1, 2022
Updated 8mo ago

Evaluation results

MethodLinks
SEERArch.=RG256, Pre-training source=uncurated data, Evaluation Protocol=linear evaluation2021.0392.6
SEERArch.=RG128, Pre-training source=uncurated data, Evaluation Protocol=linear evaluation2021.0391.6
SwAVArch.=RG128, Pre-training source=ImageNet, Evaluation Protocol=linear evaluation2021.0389.8
SupervisedArch.=RG128, Pre-training source=ImageNet, Evaluation Protocol=linear evaluation2021.0389.2
SwAVArch.=RN50, Pre-training source=ImageNet, Evaluation Protocol=linear evaluation2021.0388.9
HCLPre-train dataset=COCO, Bbox type=Ground Truth (GT), Linear evaluation=true2022.1276.51
MoCo-v2 + bboxPre-train dataset=COCO, Bbox type=Ground Truth (GT), Linear evaluation=true2022.1275.55
HCLPre-train dataset=OpenImages, Bbox type=Ground Truth (GT), Linear evaluation=true2022.1275.4
HCLPre-train dataset=OpenImages, Bbox type=Selective Search (SS), Linear evaluation=true2022.1274.31
HCLPre-train dataset=COCO, Bbox type=Selective Search (SS), Linear evaluation=true2022.1274.19
MoCo-v2 + bboxPre-train dataset=OpenImages, Bbox type=Ground Truth (GT), Linear evaluation=true2022.1273.79
MoCo-v2 + bboxPre-train dataset=COCO, Bbox type=Selective Search (SS), Linear evaluation=true2022.1273.13
MoCo-v2 + bboxPre-train dataset=OpenImages, Bbox type=Selective Search (SS), Linear evaluation=true2022.1271.82
MoCo-v2Pre-train dataset=OpenImages, Bbox type=Random crops, Linear evaluation=true2022.1269.95
MoCo-v2Pre-train dataset=COCO, Bbox type=Random crops, Linear evaluation=true2022.1264.79