Object Detection on MS COCO 2014 (val)
70.4mAP@.5ViTAEv2-S
Evaluation results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ViTAEv2-SDecoder=Cascade Mask RCNN, Schedule=3x, Params (M)=752022.02 | 70.4 | 55.6 | 51.4 | — | — | — | — | — | — | — | — | — | — | |
| DAT-TDecoder=Cascade Mask RCNN, Schedule=3x, Params (M)=86, Venue=CVPR'222022.02 | 70.1 | 55.8 | 51.3 | — | — | — | — | — | — | — | — | — | — | |
| ViTAEv2-SDecoder=Cascade Mask RCNN, Schedule=1x, Params (M)=752022.02 | 69.9 | 54.9 | 50.6 | — | — | — | — | — | — | — | — | — | — | |
| PVTv2-B2Decoder=Cascade Mask RCNN, Schedule=3x, Params (M)=83, Venue=CVMJ'222022.02 | 69.8 | 55.3 | 51.1 | — | — | — | — | — | — | — | — | — | — | |
| ViTAEv2-SDecoder=Mask RCNN, Schedule=3x, Params (M)=372022.02 | 69.4 | 52.2 | 47.8 | — | — | — | — | — | — | — | — | — | — | |
| DAT-TDecoder=Mask RCNN, Schedule=3x, Params (M)=48, Venue=CVPR'222022.02 | 69.2 | 51.6 | 47.1 | — | — | — | — | — | — | — | — | — | — | |
| Swin-TDecoder=Cascade Mask RCNN, Schedule=3x, Params (M)=86, Venue=ICCV'212022.02 | 69.2 | 54.7 | 50.4 | — | — | — | — | — | — | — | — | — | — | |
| ViTAEv2-SDecoder=Mask RCNN, Schedule=1x, Params (M)=372022.02 | 68.8 | 51 | 46.3 | — | — | — | — | — | — | — | — | — | — | |
| MViT-TDecoder=Mask RCNN, Schedule=3x, Params (M)=46, Venue=ICCV'212022.02 | 68.7 | 50.5 | 45.9 | — | — | — | — | — | — | — | — | — | — | |
| Swin-TDecoder=Mask RCNN, Schedule=3x, Params (M)=48, Venue=ICCV'212022.02 | 68.2 | 50.2 | 46 | — | — | — | — | — | — | — | — | — | — | |
| DAT-TDecoder=Cascade Mask RCNN, Schedule=1x, Params (M)=86, Venue=CVPR'222022.02 | 68.2 | 52.9 | 49.1 | — | — | — | — | — | — | — | — | — | — | |
| CrossFormer-SDecoder=Mask RCNN, Schedule=1x, Params (M)=50, Venue=ICLR'212022.02 | 68 | 49.7 | 45.4 | — | — | — | — | — | — | — | — | — | — | |
| Focal-TDecoder=Mask RCNN, Schedule=1x, Params (M)=49, Venue=NeurIPS'212022.02 | 67.7 | 49.2 | 44.8 | — | — | — | — | — | — | — | — | — | — | |
| DAT-TDecoder=Mask RCNN, Schedule=1x, Params (M)=48, Venue=CVPR'222022.02 | 67.6 | 48.5 | 44.4 | — | — | — | — | — | — | — | — | — | — | |
| PVTv2-B2Decoder=Mask RCNN, Schedule=1x, Params (M)=45, Venue=CVMJ'222022.02 | 67.1 | 49.6 | 45.3 | — | — | — | — | — | — | — | — | — | — | |
| Swin-TDecoder=Cascade Mask RCNN, Schedule=1x, Params (M)=86, Venue=ICCV'212022.02 | 67.1 | 52.2 | 48.1 | — | — | — | — | — | — | — | — | — | — | |
| Swin-TDecoder=Mask RCNN, Schedule=1x, Params (M)=47, Venue=ICCV'212022.02 | 66.6 | 47.7 | 43.7 | — | — | — | — | — | — | — | — | — | — | |
| RegionViT-BDecoder=Mask RCNN, Schedule=1x, Params (M)=92, Venue=ICLR'212022.02 | 66 | 48.2 | 44.2 | — | — | — | — | — | — | — | — | — | — | |
| PVT-SDecoder=Mask RCNN, Schedule=3x, Params (M)=44, Venue=ICCV'212022.02 | 65.3 | 46.9 | 43 | — | — | — | — | — | — | — | — | — | — | |
| ResNet-50Decoder=Cascade Mask RCNN, Schedule=3x, Params (M)=82, Venue=CVPR'162022.02 | 64.3 | 50.5 | 46.3 | — | — | — | — | — | — | — | — | — | — | |
| F-VIT+CLIPSelfbackbone=ViT-L/14, training_data=OV-LVIS2023.10 | 63.8 | 44.3 | — | 40.5 | — | — | — | — | — | — | — | — | — | |
| PVT-SDecoder=Mask RCNN, Schedule=1x, Params (M)=44, Venue=ICCV'212022.02 | 62.9 | 43.8 | 40.4 | — | — | — | — | — | — | — | — | — | — | |
| ResNet-50Decoder=Mask RCNN, Schedule=3x, Params (M)=44, Venue=CVPR'162022.02 | 61.7 | 44.9 | 41 | — | — | — | — | — | — | — | — | — | — | |
| F-VLMbackbone=ViT-L/14, training_data=OV-LVIS2023.10 | 61.6 | 43.8 | — | 39.8 | — | — | — | — | — | — | — | — | — | |
| ResNet-50Decoder=Cascade Mask RCNN, Schedule=1x, Params (M)=82, Venue=CVPR'162022.02 | 59.4 | 45 | 41.2 | — | — | — | — | — | — | — | — | — | — | |
| ResNet-50Decoder=Mask RCNN, Schedule=1x, Params (M)=44, Venue=CVPR'162022.02 | 58.6 | 41.4 | 38 | — | — | — | — | — | — | — | — | — | — | |
| BARON-KDbackbone=ViT-L/14, training_data=OV-LVIS2023.10 | 55.7 | 39.1 | — | 36.2 | — | — | — | — | — | — | — | — | — | |
| ViLDbackbone=ViT-L/14, training_data=OV-LVIS2023.10 | 55.6 | 39.8 | — | 36.6 | — | — | — | — | — | — | — | — | — | |
| DetProbackbone=ViT-L/14, training_data=OV-LVIS2023.10 | 53.8 | 37.4 | — | 34.9 | — | — | — | — | — | — | — | — | — | |
| ResNet101 + CBAMBackbone=ResNet101, Module=CBAM, Detector=Faster-RCNN2018.07 | 50.5 | 32.6 | 30.8 | — | — | — | — | — | — | — | — | — | — | |
| ResNet101Backbone=ResNet101, Detector=Faster-RCNN2018.07 | 48.4 | 30.7 | 29.1 | — | — | — | — | — | — | — | — | — | — | |
| Faster R-CNNProposal=RPN, Backbone=R101-C42020.04 | 48.4 | — | 27.2 | — | — | — | — | — | — | — | — | — | — | |
| ResNet50 + CBAMBackbone=ResNet50, Module=CBAM, Detector=Faster-RCNN2018.07 | 48.2 | 29.2 | 28.1 | — | — | — | — | — | — | — | — | — | — | |
| ResNet50Backbone=ResNet50, Detector=Faster-RCNN2018.07 | 46.2 | 28.1 | 27 | — | — | — | — | — | — | — | — | — | — | |
| MISTProposal=MCG, Backbone=R101-C42020.04 | 26.3 | — | 13 | — | — | — | — | — | — | — | — | — | — | |
| MISTProposal=MCG, Backbone=R50-C42020.04 | 26.1 | — | 12.6 | — | — | — | — | — | — | — | — | — | — | |
| MISTProposal=MCG, Backbone=VGG162020.04 | 24.3 | — | 11.4 | — | — | — | — | — | — | — | — | — | — | |
| WSOD2multi-scale testing=true2019.09 | 22.7 | — | 10.8 | — | — | — | — | — | — | — | — | — | — | |
| PCL + Fast R-CNNmulti-scale testing=true2019.09 | 19.6 | — | 9.2 | — | — | — | — | — | — | — | — | — | — | |
| PCLmulti-scale testing=true2019.09 | 19.4 | — | 8.5 | — | — | — | — | — | — | — | — | — | — | |
| Ge et al.multi-scale testing=true2019.09 | 19.3 | — | 8.9 | — | — | — | — | — | — | — | — | — | — | |
| Bency et al.2017.06 | — | — | — | 47.9 | — | — | — | — | — | — | — | — | — | |
| Conformer-S/32Decoder=Mask RCNN, Schedule=1x, Params (M)=58, Venue=ICCV'212022.02 | — | — | 43.6 | — | — | — | — | — | — | — | — | — | — | |
| ControlNetdebiasing_strategy=Bias Agnostic, detector=Faster R-CNN, backbone=ResNet-502025.10 | — | — | — | 36.9 | 37.3 | 33.4 | 27.6 | 30.8 | 38.3 | 42.9 | 49 | 40.4 | 19.8 | |
| ControlNet + Resamplingdebiasing_strategy=Frequency Aware, detector=Faster R-CNN, backbone=ResNet-502025.10 | — | — | — | 36.9 | 37.2 | 33.4 | 27.9 | 30.2 | 37.7 | 43.2 | 48.6 | 40.5 | 20.1 | |
| Copy Pastedebiasing_strategy=Bias Agnostic, detector=Faster R-CNN, backbone=ResNet-502025.10 | — | — | — | 37.9 | 38.2 | 35.5 | 28.8 | 31.4 | 39.4 | 43.6 | 48.8 | 41.5 | 21.5 | |
| Faster R-CNN (Baseline)detector=Faster R-CNN, backbone=ResNet-502025.10 | — | — | — | 37.4 | 37.7 | 33.9 | 28.3 | 31.2 | 38.9 | 43.2 | 48.1 | 41 | 21.2 | |
| GeoDiffusiondebiasing_strategy=Bias Agnostic, detector=Faster R-CNN, backbone=ResNet-502025.10 | — | — | — | 38.4 | 38.6 | 35 | 29.5 | 32 | 39.9 | 44.3 | 50.3 | 42.1 | 19.7 | |
| GeoDiffusion + Resamplingdebiasing_strategy=Frequency Aware, detector=Faster R-CNN, backbone=ResNet-502025.10 | — | — | — | 38.5 | 38.5 | 35.3 | 30 | 31.6 | 39.4 | 44.5 | 49.9 | 42.2 | 20 | |
| MSPLD2017.06 | — | — | — | 56.6 | — | — | — | — | — | — | — | — | — | |
| Oquab et al.2017.06 | — | — | — | 41.2 | — | — | — | — | — | — | — | — | — | |
| Oursdebiasing_strategy=Frequency Aware, detector=Faster R-CNN, backbone=ResNet-502025.10 | — | — | — | 40.3 | 40.5 | 36.9 | 31.5 | 33.3 | 41.8 | 46.8 | 52.5 | 43.8 | 23.1 | |
| Sun et al.2017.06 | — | — | — | 43.5 | — | — | — | — | — | — | — | — | — | |
| Zhu et al.2017.06 | — | — | — | 55.3 | — | — | — | — | — | — | — | — | — |