HyperAI超神経

プラットフォーム

Instance Segmentation On Cityscapes Val

評価指標

mask AP

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

		Paper Title
OpenSeeD( SwinL, single-scale)	49.3	A Simple Framework for Open-Vocabulary Segmentation and Detection
OneFormer (ConvNeXt-L, single-scale, Mapillary-Pretrained)	48.7	OneFormer: One Transformer to Rule Universal Image Segmentation
AFF-Base (single-scale, point-based Mask2Former)	46.2	AutoFocusFormer: Image Segmentation off the Grid
OneFormer (Swin-L, single-scale)	45.6	OneFormer: One Transformer to Rule Universal Image Segmentation
OneFormer (DiNAT-L, single-scale)	45.6	OneFormer: One Transformer to Rule Universal Image Segmentation
DiNAT-L (single-scale, Mask2Former)	45.1	Dilated Neighborhood Attention Transformer
AFF-Small (single-scale, point-based Mask2Former)	44.0	AutoFocusFormer: Image Segmentation off the Grid
Mask2Former (Swin-L, single-scale)	43.7	Masked-attention Mask Transformer for Universal Image Segmentation
Mask2Former (Swin-B)	42	Masked-attention Mask Transformer for Universal Image Segmentation
Mask2Former (Swin-S)	41.8	Masked-attention Mask Transformer for Universal Image Segmentation
PolySnake	40.2	Recurrent Generic Contour-based Instance Segmentation with Progressive Learning
Mask2Former (Swin-T)	39.7	Masked-attention Mask Transformer for Universal Image Segmentation
Mask2Former (ResNet-101)	38.5	Masked-attention Mask Transformer for Universal Image Segmentation
Mask2Former (ResNet-50)	37.4	Masked-attention Mask Transformer for Universal Image Segmentation
GAIS-Net	37.1	Geometry-Aware Instance Segmentation with Disparity Maps
PointRend	35.8	PointRend: Image Segmentation as Rendering

0 of 16 row(s) selected.

Instance Segmentation On Cityscapes Val | SOTA | HyperAI超神経