📊 Showing 5 results | 📏 Metric: mIoU
| Rank | Model | Paper | mIoU | Date | Code |
|---|---|---|---|---|---|
| 1 | Mask2Former (Swin-L) 📚 | Masked-attention Mask Transformer for Universal Image Segmentation | 47.85 | 2021-12-02 | 📦 huggingface/transformers 📦 open-mmlab/mmdetection 📦 facebookresearch/Mask2Former |
| 2 | UPerNet (ConvNeXt-L) 📚 | Unified Perceptual Parsing for Scene Understanding | 47.30 | 2018-07-26 | 📦 open-mmlab/mmsegmentation 📦 CSAILVision/semantic-segmentation-pytorch 📦 sithu31296/semantic-segmentation |
| 3 | Mask2Former (ResNet-50) 📚 | Masked-attention Mask Transformer for Universal Image Segmentation | 43.71 | 2021-12-02 | 📦 huggingface/transformers 📦 open-mmlab/mmdetection 📦 facebookresearch/Mask2Former |
| 4 | DeepLabv3 (ResNet-50) 📚 | Rethinking Atrous Convolution for Semantic Image Segmentation | 43.37 | 2017-06-17 | 📦 tensorflow/models 📦 tensorflow/models 📦 pytorch/vision |
| 5 | Segformer (MiT-B5) 📚 | SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers | 40.83 | 2021-05-31 | 📦 huggingface/transformers 📦 VikParuchuri/surya 📦 PaddlePaddle/PaddleSeg |