Instance Segmentation On Coco Minival

评估指标

APL

APM

APS

mask AP

评测结果

各个模型在此基准测试上的表现结果

					Paper Title	Repository
Co-DETR	74.6	59.7	38.9	56.6	DETRs with Collaborative Hybrid Assignments Training
ViT-CoMer-L (Mask RCNN, DINOv2)	-	-	-	55.9	ViT-CoMer: Vision Transformer with Convolutional Multi-scale Feature Interaction for Dense Predictions	-
InternImage-H	74.4	58.4	37.9	55.4	InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
EVA	72.0	58.4	37.6	55.0	EVA: Exploring the Limits of Masked Visual Representation Learning at Scale
Mask Frozen-DETR	72.9	58.4	37.2	54.9	Mask Frozen-DETR: High Quality Instance Segmentation with One GPU	-
MasK DINO (SwinL, multi-scale)	-	-	-	54.5	Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation
ViT-Adapter-L (HTC++, BEiTv2, O365, multi-scale)	-	-	-	54.2	Vision Transformer Adapter for Dense Predictions
GLEE-Pro	-	-	-	54.2	General Object Foundation Model for Images and Videos at Scale
SwinV2-G (HTC++)	-	-	-	53.7	Swin Transformer V2: Scaling Up Capacity and Resolution
ViTDet, ViT-H Cascade (multiscale)	-	-	-	53.1	Exploring Plain Vision Transformer Backbones for Object Detection
GLEE-Plus	-	-	-	53.0	General Object Foundation Model for Images and Videos at Scale
Mask DINO (SwinL)	-	-	-	52.6	Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation
ViT-Adapter-L (HTC++, BEiTv2 pretrain, multi-scale)	-	-	-	52.5	Vision Transformer Adapter for Dense Predictions
Soft Teacher + Swin-L(HTC++, multi-scale)	-	-	-	52.5	End-to-End Semi-Supervised Object Detection with Soft Teacher
ViT-Adapter-L (HTC++, BEiT pretrain, multi-scale)	-	-	-	52.2	Vision Transformer Adapter for Dense Predictions
ViTDet, ViT-H Cascade	-	-	-	52	Exploring Plain Vision Transformer Backbones for Object Detection
Soft Teacher + Swin-L(HTC++, single-scale)	-	-	-	51.9	End-to-End Semi-Supervised Object Detection with Soft Teacher
CBNetV2 (Dual-Swin-L HTC, multi-scale)	-	-	-	51.8	CBNet: A Composite Backbone Network Architecture for Object Detection
Frozen Backbone, SwinV2-G-ext22K (HTC)	-	-	-	51.6	Could Giant Pretrained Image Models Extract Universal Representations?	-
CBNetV2 (Dual-Swin-L HTC, multi-scale)	-	-	-	51	CBNet: A Composite Backbone Network Architecture for Object Detection

0 of 93 row(s) selected.