ADE20K

Introduced by Zhou et al. in Scene Parsing Through ADE20K Dataset

The ADE20K semantic segmentation dataset contains more than 20K scene-centric images exhaustively annotated with pixel-level objects and object parts labels. There are totally 150 semantic categories, which include stuffs like sky, road, grass, and discrete objects like person, car, bed.

Source: Cooperative Image Segmentation and Restoration in Adverse Environmental Conditions

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
Semantic Segmentation	ADE20K	ONE-PEACE
Semantic Segmentation	ADE20K val	BEiT-3
Panoptic Segmentation	ADE20K val	OneFormer
Open Vocabulary Semantic Segmentation	ADE20K-150	CAT-Seg
Image-to-Image Translation	ADE20K Labels-to-Photos	DP-SIMS
Open Vocabulary Semantic Segmentation	ADE20K-847	CAT-Seg
Instance Segmentation	ADE20K val	OneFormer
Unsupervised Semantic Segmentation with Language-image Pre-training	ADE20K	TagAlign
Overlapped 100-5	ADE20K	SSUL-M
Open Vocabulary Panoptic Segmentation	ADE20K	PosSAM
Image-to-Image Translation	ADE20K-Outdoor Labels-to-Photos	DP-GAN
Overlapped 100-50	ADE20K	MiB+AWT
Overlapped 50-50	ADE20K	MiB+AWT
Overlapped 100-10	ADE20K	SATS-M
Semi-Supervised Semantic Segmentation	ADE20K 1/16 labeled	SemiVL
Semi-Supervised Semantic Segmentation	ADE20K 1/32 labeled	SemiVL
Continual Semantic Segmentation	ADE20K	LGKD
Weakly-Supervised Semantic Segmentation	ADE20K val	DHR
Scene Understanding	ADE20K val	CPN
Scene Recognition	ADE20K	Semantic-Aware Scene Recogniton
Face Detection	ADE20K	CASSOD
Reconstruction	ADE20K	SCAM
Pose Transfer	ADE20K	SCAM
Overlapped 25-25	ADE20K	SATS-M
Zero-Shot Semantic Segmentation	ADE20K-847	MAFT