Semantic Segmentation On Ade20K Val

评估指标

mIoU

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
BEiT-3	62.8	Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Tasks
ViT-CoMer	62.1	ViT-CoMer: Vision Transformer with Convolutional Multi-scale Feature Interaction for Dense Predictions	-
EVA	61.5	EVA: Exploring the Limits of Masked Visual Representation Learning at Scale
FD-SwinV2-G	61.4	Contrastive Learning Rivals Masked Image Modeling in Fine-tuning via Feature Distillation
OneFormer (InternImage-H, emb_dim=256, multi-scale, 896x896)	60.8	OneFormer: One Transformer to Rule Universal Image Segmentation
MaskDINO-SwinL	60.8	Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation
ViT-Adapter-L (Mask2Former, BEiT pretrain)	60.5	Vision Transformer Adapter for Dense Predictions
SERNet-Former_v2	59.35	SERNet-Former: Semantic Segmentation by Efficient Residual Network with Attention-Boosting Gates and Attention-Fusion Networks
OneFormer (DiNAT-L, multi-scale, 896x896)	58.6	OneFormer: One Transformer to Rule Universal Image Segmentation
ViT-Adapter-L (UperNet, BEiT pretrain)	58.4	Vision Transformer Adapter for Dense Predictions
RSSeg-ViT-L(BEiT pretrain)	58.4	Representation Separation for Semantic Segmentation with Vision Transformers	-
OneFormer (DiNAT-L, multi-scale, 640x640)	58.4	OneFormer: One Transformer to Rule Universal Image Segmentation
OneFormer (Swin-L, multi-scale, 896x896)	58.3	OneFormer: One Transformer to Rule Universal Image Segmentation
SeMask (SeMask Swin-L FaPN-Mask2Former)	58.2	SeMask: Semantically Masked Transformers for Semantic Segmentation
SeMask (SeMask Swin-L MSFaPN-Mask2Former)	58.2	SeMask: Semantically Masked Transformers for Semantic Segmentation
DiNAT-L (Mask2Former)	58.1	Dilated Neighborhood Attention Transformer
Mask2Former (Swin-L-FaPN, multiscale)	57.7	Masked-attention Mask Transformer for Universal Image Segmentation
OneFormer (Swin-L, multi-scale, 640x640)	57.7	OneFormer: One Transformer to Rule Universal Image Segmentation
SeMask (SeMask Swin-L Mask2Former)	57.5	SeMask: Semantically Masked Transformers for Semantic Segmentation
SenFormer (BEiT-L)	57.1	Efficient Self-Ensemble for Semantic Segmentation

0 of 94 row(s) selected.