Referring Image Segmentation on RefCOCO+ medium (val)
68.1oIoUMagNet+
Evaluation results
| Method | Links | |
|---|---|---|
| MagNet+Backbone=Swin-B, Training Setting=Multiple Datasets2023.12 | 68.1 | |
| PolyFormerBackbone=Swin-B, Training Setting=Multiple Datasets2023.12 | 67.64 | |
| MagNetBackbone=Swin-B, Training Setting=Single Dataset2023.12 | 66.16 | |
| ReLABackbone=Swin-B, Training Setting=Single Dataset2023.12 | 66.04 | |
| MCRESBackbone=Swin-B, Training Setting=Single Dataset2023.12 | 64.32 | |
| SADLRBackbone=Swin-B, Training Setting=Single Dataset2023.12 | 64.28 | |
| VPDBackbone=Swin-B, Training Setting=Single Dataset2023.12 | 63.93 | |
| PVDBackbone=Swin-B, Training Setting=Single Dataset2023.12 | 63.38 | |
| CoupAlignBackbone=Swin-B, Training Setting=Single Dataset2023.12 | 62.92 | |
| LISA-7BBackbone=SAM-H, Training Setting=Multiple Datasets2023.12 | 62.4 | |
| CRISBackbone=ResNet-101, Training Setting=Single Dataset2023.12 | 62.27 | |
| LAVTBackbone=Swin-B, Training Setting=Single Dataset2023.12 | 62.14 | |
| ReSTRBackbone=ViT-B-16, Training Setting=Single Dataset2023.12 | 55.78 | |
| VLTBackbone=Darknet-53, Training Setting=Single Dataset2023.12 | 55.5 |