HyperAI초신경

홈 뉴스 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

2D Object Detection On Sardet 100K

평가 지표

box mAP

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	box mAP	Paper Title	Repository
MSFA (F-RCNN+R50)	51.1	SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection
Cascade R-CNN	51.1	Cascade R-CNN: Delving into High Quality Object Detection
FCOS	49.8	FCOS: Fully Convolutional One-Stage Object Detection
MSFA (F-RCNN+ConvNext-T)	54.8	SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection
DenoDet	55.4	DenoDet: Attention as Deformable Multi-Subspace Feature Denoising for Target Detection in SAR Images
Grid RCNN	48.8	Grid R-CNN
Sparse R-CNN	38.1	Sparse R-CNN: End-to-End Object Detection with Learnable Proposals
SARATR-X	-	SARATR-X: Toward Building A Foundation Model for SAR Target Recognition
Deformable DETR	50.0	Deformable DETR: Deformable Transformers for End-to-End Object Detection
MSFA (GFL+R50)	53.7	SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection
RetinaNet	47.4	Focal Loss for Dense Object Detection
F-RCNN	49.0	Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
MSFA (Deformable DETR)	51.3	SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection

0 of 13 row(s) selected.