HyperAIHyperAI超神经

Command Palette

Search for a command to run...

算力平台

Object Detection On Coco Minival

评估指标

AP50
AP75
APL
APM
APS
box AP

评测结果

各个模型在此基准测试上的表现结果

Paper Title
Co-DETR-----65.9DETRs with Collaborative Hybrid Assignments Training
M3I Pre-training (InternImage-H)-----65.0Towards All-in-one Pre-training via Maximizing Multi-modal Mutual Information
InternImage-H-----65.0InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
Co-DETR (Swin-L)-----64.7DETRs with Collaborative Hybrid Assignments Training
Focal-Stable-DINO (Focal-Huge, no TTA)81.571.478.568.550.464.6A Strong and Reproducible Object Detector with Only Public Datasets
EVA82.170.878.568.449.464.5EVA: Exploring the Limits of Masked Visual Representation Learning at Scale
ViT-CoMer-----64.3ViT-CoMer: Vision Transformer with Convolutional Multi-scale Feature Interaction for Dense Predictions
FocalNet-H (DINO)-----64.2Focal Modulation Networks
InternImage-XL-----64.2InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions
CP-DETR-L Swin-L(Fine tuning separately in COCO)-----64.1CP-DETR: Concept Prompt Guide DETR Toward Stronger Universal Object Detection
RevCol-H(DINO)-----63.8Reversible Column Networks
DINO (Swin-L)-----63.2DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection
Grounding DINO-----63.0Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
SwinV2-G (HTC++)-----62.5Swin Transformer V2: Scaling Up Capacity and Resolution
GLEE-Pro-----62.0General Object Foundation Model for Images and Videos at Scale
Florence-CoSwin-H-----62Florence: A New Foundation Model for Computer Vision
ViTDet, ViT-H Cascade (multiscale)-----61.3Exploring Plain Vision Transformer Backbones for Object Detection
GLIP (Swin-L, multi-scale)-----60.8Grounded Language-Image Pre-training
Soft Teacher + Swin-L (HTC++, multi-scale)-----60.7End-to-End Semi-Supervised Object Detection with Soft Teacher
UNINEXT-H77.566.775.364.845.160.6Universal Instance Perception as Object Discovery and Retrieval
0 of 219 row(s) selected.
Object Detection On Coco Minival | SOTA | HyperAI超神经