Interactive Image Segmentation on GrabCut
1.36NoC@90InterFormer
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| InterFormerTrain data=COCO+LVIS, Backbone=VIT-L, SPC/s=0.271, Fusion Strategy=Late Fusion2024.08 | 1.36 | — | — | |
| SAM-REFTrain data=COCO+LVIS, Backbone=ViT-H*, SPC/s=0.511, Fusion Strategy=Late Fusion2024.08 | 1.36 | — | 2.2 | |
| SAM-REFTrain data=COCO+LVIS, Backbone=ViT-B*, SPC/s=0.278, Fusion Strategy=Late Fusion2024.08 | 1.4 | — | 2.44 | |
| iCMFormerBackbone=ViT-B2024.04 | 1.42 | 1.36 | — | |
| FocSAMTrain data=COCO+LVIS, Backbone=VIT-H*, SPC/s=0.467, Fusion Strategy=Late Fusion2024.08 | 1.44 | — | 2.6 | |
| MFPBackbone=ViT-B2024.04 | 1.48 | 1.38 | 1.92 | |
| PseudoClickTrain data=COCO+LVIS, Backbone=HRNet32, Fusion Strategy=Early Fusion2024.08 | 1.5 | — | — | |
| SimpleClickTrain data=COCO+LVIS, Backbone=VIT-H, SPC/s=8.24, Fusion Strategy=Early Fusion2024.08 | 1.5 | — | 1.66 | |
| InterFormerTrain data=COCO+LVIS, Backbone=VIT-B, SPC/s=0.153, Fusion Strategy=Late Fusion2024.08 | 1.5 | — | — | |
| CDNetTrain data=COCO+LVIS, Backbone=ResNet, Fusion Strategy=Early Fusion2024.08 | 1.52 | — | — | |
| SimpleClickBackbone=ViT-B2024.04 | 1.54 | 1.4 | 2.16 | |
| RITMTrain data=COCO+LVIS, Backbone=HRNet32, Fusion Strategy=Early Fusion2024.08 | 1.56 | — | 2.48 | |
| FocusCutBackbone=ResNet-1012024.04 | 1.64 | 1.46 | — | |
| FocalClickTrain data=COCO+LVIS, Backbone=SegFB3-S2384, SPC/s=0.232, Fusion Strategy=Early Fusion2024.08 | 1.68 | — | 1.92 | |
| AdaptiveClickTrain data=COCO+LVIS, Backbone=VIT-H, Fusion Strategy=Early Fusion2024.08 | 1.69 | — | — | |
| EdgeFlowTrain data=COCO+LVIS, Backbone=HRNet18, Fusion Strategy=Early Fusion2024.08 | 1.72 | — | — | |
| SAM-REFTrain data=HQSeg-44K, Backbone=ViT-H*, SPC/s=0.511, Fusion Strategy=Late Fusion2024.08 | 1.76 | — | 2.22 | |
| GPCISBackbone=ResNet-502024.04 | 1.82 | 1.64 | 2.62 | |
| EMC-ClickBackbone=HRNet-182024.04 | 1.84 | 1.74 | — | |
| HQ-SAMTrain data=HQSeg-44K, Backbone=ViT-H*, SPC/s=0.443, Fusion Strategy=Late Fusion2024.08 | 1.86 | — | 2.26 | |
| SAMTrain data=SA-1B, Backbone=ViT-H, SPC/s=0.413, Fusion Strategy=Late Fusion2024.08 | 1.88 | — | 2.28 | |
| FocalClickBackbone=SegF-BO2024.04 | 1.9 | 1.66 | — | |
| RITMBackbone=HRNet-182024.04 | 2.04 | 1.76 | 3.66 | |
| PseudoClickBackbone=HRNet-182024.04 | 2.04 | 1.68 | — | |
| CDNetBackbone=ResNet-342024.04 | 2.18 | 1.86 | 3.68 | |
| f-BRSBackbone=ResNet-1012024.04 | 2.72 | 2.3 | — | |
| F-BRSTrain data=COCO+LVIS, Backbone=ResNet, Fusion Strategy=Early Fusion2024.08 | 2.72 | — | — | |
| BRSBackbone=DenseNet2024.04 | 3.6 | 2.6 | — |