Semantic Segmentation On Ade20K

Métriques

GFLOPs

Params (M)

Validation mIoU

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	GFLOPs	Params (M)	Validation mIoU	Paper Title	Repository
InternImage-L	2526	256	54.1	InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
TEC (Vit-B, Upernet)	-	-	51.0	Towards Sustainable Self-supervised Learning
EfficientViT-B3 (r512)	-	-	49	EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Prediction
VAN-Large	-	49	48.1	Visual Attention Network
MaskFormer(Swin-B)	-	-	53.8	Per-Pixel Classification is Not All You Need for Semantic Segmentation
CFNet(ResNet-101)	-	-	44.89	Co-Occurrent Features in Semantic Segmentation
ConvNeXt-S	-	82	49.6	A ConvNet for the 2020s
SegNet	-	-	21.64	SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
M3I Pre-training (InternImage-H)	-	1310	62.9	Towards All-in-one Pre-training via Maximizing Multi-modal Mutual Information
HRViT-b2 (SegFormer, SS)	-	20.8	48.76	Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation
SwinV2-G-HTC++ Liu et al. ([2021a])	-	-	53.7	Swin Transformer V2: Scaling Up Capacity and Resolution
SETR-MLA (160k, MS)	-	-	50.28	Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
ELSA-Swin-S	-	-	50.3	ELSA: Enhanced Local Self-Attention for Vision Transformer
Sequential Ensemble (DeepLabv3+)	-	-	46.8	Sequential Ensembling for Semantic Segmentation	-
FastViT-SA36	-	-	-	FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization
ACNet (ResNet-101)	-	-	45.90	Adaptive Context Network for Scene Parsing	-
SGR (ResNet-101)	-	-	44.32	Symbolic Graph Reasoning Meets Convolutions
ConvMLP-S	-	-	35.8	ConvMLP: Hierarchical Convolutional MLPs for Vision
SeMask (SeMask Swin-L MSFaPN-Mask2Former)	-	-	58.2	SeMask: Semantically Masked Transformers for Semantic Segmentation
ConvMLP-M	-	-	38.6	ConvMLP: Hierarchical Convolutional MLPs for Vision

0 of 230 row(s) selected.