Instance Segmentation on ADE20K val

Metric: APS (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Hide extra data

Sort:

#	Model↕	APS▼	Extra Data	Paper	Date↕	Code
1	OneFormer (InternImage-H, emb_dim=1024, single-scale, 896x896, COCO-Pretrained)	23.7	No	OneFormer: One Transformer to Rule Universal Ima...	2022-11-10	Code
2	OneFormer (DiNAT-L, single-scale, 1280x1280, COCO-pretrain)	19.2	Yes	OneFormer: One Transformer to Rule Universal Ima...	2022-11-10	Code
3	X-Decoder (Davit-d5, Deform, single-scale, 1280x1280)	18.9	Yes	Generalized Decoding for Pixel, Image, and Langu...	2022-12-21	Code
4	DiNAT-L (Mask2Former, single-scale)	16.3	No	Dilated Neighborhood Attention Transformer	2022-09-29	Code
5	Mask2Former (Swin-L, single-scale)	16.3	No	Masked-attention Mask Transformer for Universal ...	2021-12-02	Code
6	Mask2Former (Swin-L + FAPN)	14.6	No	Masked-attention Mask Transformer for Universal ...	2021-12-02	Code
7	Mask2Former (ResNet50)	10.4	No	Masked-attention Mask Transformer for Universal ...	2021-12-02	Code

#1OneFormer (InternImage-H, emb_dim=1024, single-scale, 896x896, COCO-Pretrained)SOTA
23.7
APS· 2022-11-10
OneFormer: One Transformer to Rule Universal Image Segmentation Code
#2OneFormer (DiNAT-L, single-scale, 1280x1280, COCO-pretrain)
19.2
APS· Extra Data· 2022-11-10
OneFormer: One Transformer to Rule Universal Image Segmentation Code
#3X-Decoder (Davit-d5, Deform, single-scale, 1280x1280)
18.9
APS· Extra Data· 2022-12-21
Generalized Decoding for Pixel, Image, and Language Code
#4DiNAT-L (Mask2Former, single-scale)
16.3
APS· 2022-09-29
Dilated Neighborhood Attention Transformer Code
#5Mask2Former (Swin-L, single-scale)SOTA
16.3
APS· 2021-12-02
Masked-attention Mask Transformer for Universal Image Segmentation Code
#6Mask2Former (Swin-L + FAPN)
14.6
APS· 2021-12-02
Masked-attention Mask Transformer for Universal Image Segmentation Code
#7Mask2Former (ResNet50)
10.4
APS· 2021-12-02
Masked-attention Mask Transformer for Universal Image Segmentation Code