Object Detection on VOC 2007 (test)
85.4AP@50BYOL
Evaluation results
| Method | Links | ||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| BYOLEvaluation Protocol=Fine-tuned2021.06 | 85.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Supervised-INEvaluation Protocol=Fine-tuned2021.06 | 85 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RetinaNet500+AP-Loss (ours)Backbone=ResNet-101, Multi-Scale=true2019.04 | 84.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AP-loss500+Backbone=ResNet-101, Stage=One-Stage, multi-scale=true2020.08 | 84.9 | — | — | — | — | — | 88.9 | 89.6 | 87.8 | 81.7 | 76.2 | 89 | 89.5 | 89.8 | 74.8 | 87.9 | 79.3 | 88.6 | 89.8 | 88.7 | 87.7 | 66.5 | 86.7 | 84.2 | 88.4 | 85.4 | |
| SimCLREvaluation Protocol=Fine-tuned2021.06 | 84.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSL-HSIC (w/o target)Evaluation Protocol=Fine-tuned2021.06 | 84.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSL-HSIC (w/ target)Evaluation Protocol=Fine-tuned2021.06 | 84.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PFPNet-R512Backbone=VGG-16, Multi-Scale=true2019.04 | 84.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PFPNet-R512+Backbone=VGG-16, Stage=One-Stage, multi-scale=true2020.08 | 84.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Revisiting-RCNNBackbone=ResNet-101+152, Stage=Two-Stage, multi-scale=false2020.08 | 84 | — | — | — | — | — | 89.3 | 88.7 | 80.5 | 77.7 | 76.3 | 90.1 | 89.6 | 89.8 | 72.9 | 89.2 | 77.8 | 90.1 | 90 | 87.5 | 87.2 | 58.6 | 88.2 | 84.3 | 87.5 | 85 | |
| RetinaNet500+AP-Loss (ours)Backbone=ResNet-101, Multi-Scale=false2019.04 | 83.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AP-loss500Backbone=ResNet-101, Stage=One-Stage, multi-scale=false2020.08 | 83.9 | — | — | — | — | — | 87.2 | 88.3 | 85.9 | 80.5 | 73.6 | 87.9 | 89.5 | 89.8 | 71.6 | 88.8 | 77.4 | 88.8 | 89.8 | 89.3 | 87 | 63.3 | 86.6 | 81.5 | 87.8 | 83.1 | |
| RefineDet512Backbone=VGG-16, Multi-Scale=true2019.04 | 83.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RefineDet512+Backbone=VGG-16, Stage=One-Stage, multi-scale=true2020.08 | 83.8 | — | — | — | — | — | 88.5 | 89.1 | 85.5 | 79.8 | 72.4 | 89.5 | 89.5 | 89.9 | 69.9 | 88.9 | 75.9 | 87.4 | 89.6 | 89 | 86.2 | 63.9 | 86.2 | 81 | 88.6 | 84.4 | |
| PixProPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 83.4 | 59.5 | 66.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSL-HSIC (w/ target)Evaluation Protocol=Linear2021.06 | 83.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReSimPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 83.1 | 58.7 | 66.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OBoWEpochs=200, Batch=256, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 82.9 | — | 64.8 | 57.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DenseSiamPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 82.9 | 58.5 | 65.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Supervised-INEvaluation Protocol=Linear2021.06 | 82.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DenseCLPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 82.8 | 58.7 | 65.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CoupleNetBackbone=ResNet-101, Stage=Two-Stage, multi-scale=false2020.08 | 82.7 | — | — | — | — | — | 85.7 | 87 | 84.8 | 75.5 | 73.3 | 88.8 | 89.2 | 89.6 | 69.8 | 87.5 | 76.1 | 88.9 | 89 | 87.2 | 86.2 | 59.1 | 83.6 | 83.4 | 87.6 | 80.7 | |
| SwAVEpochs=800, Batch=4096, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 82.6 | — | 62.7 | 56.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DetCoPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 82.6 | 57.8 | 64.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MoCo v2Epochs=800, Batch=256, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 82.5 | — | 64 | 57.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BYOLEvaluation Protocol=Linear2021.06 | 82.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MoCo v2Epochs=200, Batch=256, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 82.4 | — | 63.6 | 57 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MoCo v2Pre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 82.4 | 57 | 63.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimSiamPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 82.3 | 56.7 | 63.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PFPNet-R512Backbone=VGG-16, Multi-Scale=false2019.04 | 82.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PFPNet-R512Backbone=VGG-16, Stage=One-Stage, multi-scale=false2020.08 | 82.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RFBNet512Backbone=VGG-16, Multi-Scale=false2019.04 | 82.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RFBNet512Backbone=VGG-16, Stage=One-Stage, multi-scale=false2020.08 | 82.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SwAVEpochs=200, Batch=256, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 81.8 | — | 60 | 54.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RefineDet512Backbone=VGG-16, Multi-Scale=false2019.04 | 81.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RefineDet512Backbone=VGG-16, Stage=One-Stage, multi-scale=false2020.08 | 81.8 | — | — | — | — | — | 88.7 | 87 | 83.2 | 76.5 | 68 | 88.5 | 88.7 | 89.2 | 66.5 | 87.9 | 75 | 86.8 | 89.2 | 87.8 | 84.7 | 56.2 | 83.2 | 78.7 | 88.1 | 82.3 | |
| DES512Backbone=VGG-16, Multi-Scale=false2019.04 | 81.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DES512Backbone=VGG-16, Stage=One-Stage, multi-scale=false2020.08 | 81.7 | — | — | — | — | — | 87.7 | 86.7 | 85.2 | 76.3 | 60.6 | 88.7 | 89 | 88 | 67 | 86.9 | 78 | 87.2 | 87.9 | 87.4 | 84.4 | 59.2 | 86.1 | 79.2 | 88.1 | 80.5 | |
| supervisedPre-train=ImageNet supervised, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 81.6 | 54.2 | 59.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DSSD513Backbone=ResNet-101, Multi-Scale=false2019.04 | 81.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DSSD513Backbone=ResNet-101, Stage=One-Stage, multi-scale=false2020.08 | 81.5 | — | — | — | — | — | 86.6 | 86.2 | 82.6 | 74.9 | 62.5 | 89 | 88.7 | 88.8 | 65.2 | 87 | 78.7 | 88.2 | 89 | 87.5 | 83.7 | 51.1 | 86.3 | 81.6 | 85.7 | 83.7 | |
| SSL-HSIC (w/o target)Evaluation Protocol=Linear2021.06 | 81.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SupervisedEpochs=100, Batch=256, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 81.3 | — | 58.8 | 53.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BoWNetEpochs=325, Batch=256, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 81.3 | — | 61.1 | 55.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BYOLPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 81 | 51.9 | 56.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PIRLEpochs=800, Batch=1024, Evaluation Protocol=Fine-tuning, Backbone=ResNet-50, Detector=Faster R-CNN (R50-C4)2020.12 | 80.7 | — | 59.7 | 54 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSD513Backbone=ResNet-101, Multi-Scale=false2019.04 | 80.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSD513Backbone=ResNet-101, Stage=One-Stage, multi-scale=false2020.08 | 80.6 | — | — | — | — | — | 84.3 | 87.6 | 82.6 | 71.6 | 59 | 88.2 | 88.1 | 89.3 | 64.4 | 85.6 | 76.2 | 88.5 | 88.9 | 87.5 | 83 | 53.6 | 83.9 | 82.2 | 87.2 | 81.3 | |
| SimCLREvaluation Protocol=Linear2021.06 | 80.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| R-FCNBackbone=ResNet-101, Stage=Two-Stage, multi-scale=false2020.08 | 80.5 | — | — | — | — | — | 79.9 | 87.2 | 81.5 | 72 | 69.8 | 86.8 | 88.5 | 89.8 | 67 | 88.1 | 74.5 | 89.8 | 90.6 | 79.9 | 81.2 | 53.7 | 81.8 | 81.5 | 85.9 | 79.9 | |
| YOLOv3training set=normal, testing set=normal, pre-process=-2022.05 | 80.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSD512Backbone=VGG-16, Multi-Scale=false2019.04 | 79.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSD512*Backbone=VGG-16, Stage=One-Stage, multi-scale=false2020.08 | 79.8 | — | — | — | — | — | 84.8 | 85.1 | 81.5 | 73 | 57.8 | 87.8 | 88.3 | 87.4 | 63.5 | 85.4 | 73.2 | 86.2 | 86.7 | 83.9 | 82.5 | 55.6 | 81.7 | 79 | 86.6 | 80 | |
| SimCLRPre-train=200-epoch IN1k, Backbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 79.4 | 51.5 | 55.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAET (w ort)training set=low+normal, testing set=low, pre-process=-2022.05 | 78.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| YOLOv2Backbone=DarkNet-19, Multi-Scale=false2019.04 | 78.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| YOLOv2Backbone=DarkNet-19, Stage=One-Stage, multi-scale=false2020.08 | 78.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MR-CNNBackbone=VGG-16, Stage=Two-Stage, multi-scale=false2020.08 | 78.2 | — | — | — | — | — | 80.3 | 84.1 | 78.5 | 70.8 | 68.5 | 88 | 85.9 | 87.8 | 60.3 | 85.2 | 73.7 | 87.2 | 86.5 | 85 | 76.4 | 48.5 | 76.3 | 75.5 | 85 | 81 | |
| DSOD300Backbone=DS/64-192-48-1, Multi-Scale=false2019.04 | 77.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DSOD300Backbone=DS/64-192-48-1, Stage=One-Stage, multi-scale=false2020.08 | 77.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAET (w/o ort)training set=low+normal, testing set=low, pre-process=-2022.05 | 77 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| IONBackbone=VGG-16, Stage=Two-Stage, multi-scale=false2020.08 | 76.5 | — | — | — | — | — | 79.2 | 79.2 | 77.4 | 69.8 | 55.7 | 85.2 | 84.2 | 89.8 | 57.5 | 78.5 | 73.8 | 87.8 | 85.9 | 81.3 | 75.3 | 49.7 | 76.9 | 74.6 | 85.2 | 82.1 | |
| YOLOv3training set=low, testing set=low, pre-process=-2022.05 | 76.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FasterBackbone=ResNet-101, Stage=Two-Stage, multi-scale=false2020.08 | 76.4 | — | — | — | — | — | 79.8 | 80.7 | 76.2 | 68.3 | 55.9 | 85.1 | 85.3 | 89.8 | 56.7 | 87.8 | 69.4 | 88.3 | 88.9 | 80.9 | 78.4 | 41.7 | 78.6 | 79.8 | 85.3 | 72 | |
| YOLOv3training set=normal, testing set=low, pre-process=KIND2022.05 | 72.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| YOLOv3training set=normal, testing set=low, pre-process=Zero-DCE2022.05 | 71.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| YOLOv3training set=normal, testing set=low, pre-process=MBLLEN2022.05 | 71.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Random initEvaluation Protocol=Fine-tuned2021.06 | 67.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SoS-WSODBackbone=ResNet50, Stages=stage 1+2+3, Annotation Level=Pure WSOD, TTA=true2021.06 | 64.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SoS-WSODBackbone=VGG16, Stages=stage 1+2+3, Annotation Level=Pure WSOD, TTA=true2021.06 | 60.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OCUD + FRBackbone=ResNet50, Annotation Level=WSOD with transfer (COCO-60), TTA=true2021.06 | 60.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ada2MSOptimization algorithm=Ada2MS2026.05 | 59.4 | — | 35.46 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CaTBackbone=VGG16, Annotation Level=WSOD with transfer (COCO-60), TTA=true2021.06 | 59.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AdamWOptimization algorithm=AdamW2026.05 | 59.12 | — | 34.91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| scratchBackbone=ResNet-50-C4, Model=Faster R-CNN2022.03 | 59 | 32.8 | 31.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RAdamOptimization algorithm=RAdam2026.05 | 58.89 | — | 34.66 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Momentum SGDOptimization algorithm=Momentum SGD2026.05 | 57.19 | — | 31.02 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LionOptimization algorithm=Lion2026.05 | 56.97 | — | 32.14 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AdaIOptimization algorithm=AdaI2026.05 | 56.91 | — | 31.31 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CASDBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 56.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LBBABackbone=VGG16, Annotation Level=WSOD with transfer (COCO-60), TTA=true2021.06 | 56.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SoS-WSODBackbone=VGG16, Stages=stage 1, Annotation Level=Pure WSOD, TTA=true2021.06 | 55 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MISTBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 54.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SophiaGOptimization algorithm=SophiaG2026.05 | 54.61 | — | 29.07 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| IM-CFBBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 54.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SLV + FRBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 53.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| C-MIDN + FRBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 53.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| WSOD2Backbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 53.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Pred NetBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 52.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| W2FBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 52.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PCLBackbone=VGG16, Annotation Level=Pure WSOD, TTA=true2021.06 | 43.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| D-YOLOBackbone=Tiny Darknet, Input=416 x 416, MFLOPS=20902019.03 | — | 67.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Dropout Sampling2021.03 | — | 78.15 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DS-MBNet-L + FSSDMAdds=3.2B2021.03 | — | 73.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DS-MBNet-M + FSSDMAdds=2.7B2021.03 | — | 72.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DS-MBNet-S + FSSDMAdds=2.3B2021.03 | — | 70.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DSSD321Backbone=ResNet-101 + FPN, Input=321 x 321, MFLOPS=212002019.03 | — | 78.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EdgeBoxesTraining set=N/A2021.09 | — | 4.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FSDNet=RES-101-VOC07, Supervision=Fully Supervised2018.10 | — | 75 | — | — | — | — | 73.6 | 82.3 | 75.4 | 64 | 57.4 | 80.2 | 86.5 | 86.2 | 52.7 | 85.2 | 66.9 | 87 | 87.1 | 82.9 | 81.2 | 45.7 | 76.8 | 71.2 | 82.6 | 75.5 | |
| FSDNet=RES-152-COCO, Supervision=Fully Supervised2018.10 | — | 82.7 | — | — | — | — | 91 | 90.4 | 88.3 | 61.2 | 77.7 | 92.2 | 82.2 | 93.2 | 67 | 89.4 | 65.8 | 88 | 92 | 89.5 | 88.5 | 56.9 | 85.1 | 81 | 89.8 | 85.2 |