Object Detection In Aerial Images On Dota 1

Métriques

mAP

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	mAP	Paper Title	Repository
EAutoDet	77.05%	EAutoDet: Efficient Architecture Search for Object Detection	-
RSP-ViTAEv2-S-FPN-ORCN	77.72%	An Empirical Study of Remote Sensing Pretraining
IMP+MTP(InternImage-XL)	80.77%	MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining
ViT-B + RVSA-ORCN	81.01%	Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
ViTAE-B + RVSA-ORCN	81.24%	Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model
LSKNet-T	81.37%	LSKNet: A Foundation Lightweight Backbone for Remote Sensing
ARC	81.77%	Adaptive Rotated Convolution for Rotated Object Detection
CFC-NET	73.50%	CFC-Net: A Critical Feature Capturing Network for Arbitrary-Oriented Object Detection in Remote Sensing Images
DecoupleNet D2	78.04%	DecoupleNet: A Lightweight Backbone Network With Efficient Feature Decoupling for Remote Sensing Visual Tasks
Oriented RepPoints	77.63%	Oriented RepPoints for Aerial Object Detection
ReDet	80.10%	ReDet: A Rotation-equivariant Detector for Aerial Object Detection
DLA+S2A-Net	76.95%	Dynamic Anchor Learning for Arbitrary-Oriented Object Detection
DRN	73.23%	Dynamic Refinement Network for Oriented and Densely Packed Object Detection
Oriented RCNN	80.87%	Oriented R-CNN for Object Detection
SCRDet	72.61%	SCRDet: Towards More Robust Detection for Small, Cluttered and Rotated Objects
APE	75.75	Adaptive Period Embedding for Representing Oriented Objects in Aerial Images	-
LSKNet-S*	81.85%	LSKNet: A Foundation Lightweight Backbone for Remote Sensing
RoI Transformer	69.56%	Learning RoI Transformer for Detecting Oriented Objects in Aerial Images
CFA	76.67%	Beyond Bounding-Box: Convex-Hull Feature Adaptation for Oriented and Densely Packed Object Detection
GWD+R3Det	80.23%	Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Loss

0 of 58 row(s) selected.