Skip to content

Latest commit

 

History

History
288 lines (273 loc) · 137 KB

README.md

File metadata and controls

288 lines (273 loc) · 137 KB
Raw

DOTA-DOAI

Abstract

This repo is the codebase for our team to participate in DOTA related competitions, including rotation and horizontal detection. We mainly use FPN-based two-stage detector, and it is completed by Xue Yang and Jirui Yang.

We also recommend two rotated/oriented object detection benchmarks, which are led by Xue Yang and Yue Zhou.

Performance

DOTA1.0 (Task1)

Model Backbone Training data Val data mAP Model Link Tricks lr schd Data Augmentation GPU Image/GPU Configs
FPN ResNet152_v1d (600,800,1024)->MS DOTA1.0 trainval DOTA1.0 test 78.99 model ALL 2x Yes 2X GeForce RTX 2080 Ti 1 cfgs_dota1.0_res152_v1.py

DOTA1.0 (Task2)

Model Backbone Training data Val data mAP Model Link Tricks lr schd Data Augmentation GPU Image/GPU Configs
FPN (memory consumption) ResNet152_v1d (600,800,1024)->MS DOTA1.0 trainval DOTA1.0 test 81.23 model ALL 2x Yes 2X Quadro RTX 8000 1 cfgs_dota1.0_res152_v1.py

Visualization

1

Performance of published papers on DOTA datasets

DOTA1.0 (Task1)

HBox-supervised Model Backbone mAP Paper Link Code Link Remark Recommend
H2RBox ResNet50 74.40 ICLR2023 Pytorch, Jittor weak- and self-supervised ✅
RBox-supervised Model Backbone mAP Paper Link Code Link Remark Recommend
FR-O (DOTA) ResNet101 52.93 CVPR2018 MXNet DOTA dataset, baseline ✅
IENet ResNet101 57.14 arXiv:1912.00969 - anchor free
TOSO ResNet101 57.52 ICASSP2020 - geometric transformation
Wang et al. HRNet 58.29 Comput. Electr. Eng. - anchor free
PIoU Loss DLA-34 60.5 ECCV2020 PyTorch IoU loss, anchor free ✅
R2CNN ResNet101 60.67 arXiv:1706.09579 TF scene text, multi-task, different pooled sizes, baseline ✅
RRPN ResNet101 61.01 TMM arXiv:1703.01086 TF scene text, rotation proposals, baseline ✅
Axis Learning ResNet101 65.98 Remote Sensing Pytorch single stage, anchor free ✅
Li et al. ResNet50 66.01 IGARSS2021 - refine, feature alignment
MARNet ResNet101 67.19 IJRS - based on scrdet
ICN ResNet101 68.16 ACCV2018 - image cascade, multi-scale ✅
GSDet ResNet101 68.28 TIP - scale reasoning
RADet ResNeXt101 69.09 Remote Sensing - enhanced FPN, mask rcnn
KARNET ResNet50 68.87 CISNRC 2020 - attention denoising, anchor refining
RoI Transformer ResNet101 69.56 CVPR2019 MXNet, PyTorch, MMRotate roi transformer ✅
CAD-Net ResNet101 69.90 TGRS arXiv:1903.00857 - attention
ProbIoU ResNet50 70.04 arXiv:2106.06072 TF gaussian bounding boxes, hellinger distance ✅
ROTP ResNet101 70.29 arXiv:2202.06565 - solar corona heatmap, key points, head of object
A2S-Det ResNet101 70.64 Remote Sensing - label assign
AOOD ResNet101 71.18 Neural Computing and Applications - attention + R-DFPN
CGP Box ResNet18 71.35 IJRS - center-guide points
Zhou et al. CSPDarknet53 71.5 ijgi - enhanced fpn
ACE DLA34 71.7 TIP - corner point ✅
Pei et al. ResNet101 71.76 IGRASS2021 - enhanced FPN
Cascade-FF ResNet152 71.80 ICME2020 - refined retinanet + feature fusion
SCPNet Hourglass104 72.20 GRSL - corner points
P-RSDet ResNet101 72.30 Access - anchor free, polar coordinates ✅
Zhang et al. ResNet101 72.37 GSIS - refine-stage
ROPDet ResNet101-DCN 72.42 J REAL-TIME IMAGE PR - point set representation
SCRDet ResNet101 72.61 ICCV2019 TF: R2CNN++, IoU-Smooth L1: RetinaNet-based, R3Det-based attention, angular boundary problem ✅
O2-DNet Hourglass104 72.8 ISPRS, arXiv:1912.10694 - centernet, anchor free ✅
HRPNet HRNet-W48 72.83 GRSL - polar
SARD ResNet101 72.95 Access - IoU-based weighted loss
GLS-Net ResNet101 72.96 Remote Sensing - attention, saliency pyramid
ProjBB ResNet101 73.03 Access code, codebase new definition of bounding box
DRN Hourglass104 73.23 CVPR2020 code centernet, feature selection module, dynamic refinement head, new dataset (SKU110K-R) ✅
FADet ResNet101 73.28 ICIP2019 - attention
RBA-CenterNet ResNet101 73.41 IJCNN - centernet, refine feature
MFIAR-Net ResNet152 73.49 Sensors - feature attention, enhanced FPN
CFC-NET ResNet101 73.50 TGRS PyTorch critical feature, label assign, refine ✅
Dual-Det ResNet18 73.62 IJRS code keypoint-based
Li et al. CSP-Hourglass 73.70 GRSL - CSP-Hourglass Net
R3Det ResNet101 73.79 AAAI2021 TF, r3det-on-mmdetection, r3det-pytorch, MMRotate refined single stage, feature alignment ✅
SDCDet ResNet101 73.89 PRAI2021 - instance segmentation direction correction
Geng et al. ResNet101 73.92 GRSL - anchor free, angle encoding
Free3Net ResNet101 74.04 TMM - anchor free, gliding vertex ✅
SegmRDet ResNet50 74.14 Neurocomputing - segmentation-baed, new training and inference
Hou et al. ResNet101 74.44 TIP - enhanced FPN, feature alignment
Wu et al. ResNet50 74.45 J Electron Imaging - enhanced FPN, feature alignment
CenterRot ResNet152 74.75 Remote Sensing - anchor free, deformable-fpn, csl
MEAD ResNet101 74.80 Applied Intelligence - mechanism anchor free, mask guided, refine feature
TS4Net ResNet101 74.82 Neurocomputing - label assign
FEDet ResNet50 74.89 ICCSE 2021 - refine feature, angle constraint
SRep-RDet RepVGG-B1g2 74.89 IJRS - refine feature, attention, repvgg
Yuan et al. HRNet32 74.97 ICCTIT2021 - centernet-r
Gliding Vertex ResNet101 75.02 TPAMI arXiv:1911.09358 PyTorch, MMRotate quadrilateral bbox ✅
OSSDet ResNeXt101 75.08 JSTARS - feature enhancement and alignment
LO-Det Darknet53 75.24 TGRS, arXiv:2209.07709 PyTorch lightweight ✅
EFN U-Net 75.27 Preprints - Field-based ✅
SAR ResNet152 75.26 Access - boundary problem ✅
TricubeNet Hourglass104 75.26 WACV2022 code 2D tricube kernel ✅
Mask OBB ResNeXt101 75.33 Remote Sensing - attention, multi-task ✅
BBAVectors ResNet101 75.36 WACV2021 PyTorch keypoint based ✅
SAOA ResNet101 75.41 ICIG2021 - anchor free, spatial self-attention
Yuan et al. ResNet101 75.43 TGRS - feature alignment, continuous boundary
Zand et al. DarkNet53 75.5 TGRS - angle classification
MSFF - 75.60 ICCECE 2022 - multi-scale feature fusion
RODFormer ViT-B4 75.60 Sensors - vision transformer
FFA ResNet101 75.7 ISPRS - enhanced FPN, rotation proposals
CBDA-Net DLA-34-DCN 75.74 TGRS - dual attention
APE ResNeXt101(32x4) 75.75 TGRS arXiv:1906.09447 - adaptive period embedding, length independent IoU (LIIoU) ✅
R4Det ResNet152 75.54 Image Vis Comput - feature recursion and refinement
SurroundNet ResNet152 75.88 Remote Sensing - attention, anchor free, quad
RIE HRGANet-W48 75.94 Remote Sensing - center-based rotated inscribed ellipse
F3-Net ResNet152 76.02 Remote Sensing - feature fusion and filtration
CenterMap OBB ResNet101 76.03 TGRS - center-probability-map
DA-Net ResNet101 76.11 GRSL - feature alignment
MDL-p ResNet101 76.16 arXiv:2204.00840 - gaussian modeling
CSL ResNet152 76.17 ECCV2020 TF, MMRotate, Pytorch: YOLOv5_DOTA_OBB (CSL) angular boundary problem ✅
MRDet ResNet101 76.24 TGRS - arbitrary-oriented rpn, multiple subtasks
AFC-Net ResNet101 76.27 Neurocomputing - adaptive feature concatenate
RSDet/RSDet++ ResNet152 76.30 AAAI2021/TCSVT TF quadrilateral bbox, angular boundary problem ✅
OWSR Ensemble 76.36 CVPR2019 WorkShop - enhanced FPN
SLA ResNet50 76.36 Remote Sensing PyTorch sparse label assignment ✅
SE2-Det ResNet101 76.42 Remote Sensing - enhanced fpn
OPLD ResNet101 76.43 J-STARS PyTorch boundary problem, point-guided ✅
Polar Ray ResNet101 76.50 ACM MM2021 - polar rays representation ✅
SIoU ResNet50 76.54 JSTARS - splicing intersection over union ✅
R3Det++ ResNet152 76.56 arXiv:2004.13316 TF refined single stage, feature alignment, denoising ✅
PolarDet ResNet101 76.64 IJRS arXiv:2010.08720 - polar, center-semantic ✅
Beyond Bounding-Box ResNet152 76.67 CVPR2021 PyTorch, MMRotate point-based, reppoints ✅
OAN ResNeX50 76.73 arXiv:2212.13136 PyTorch objectness activation network, efficient ✅
SCRDet++ ResNet101 76.81 TPAMI TF angular boundary problem, denoising ✅
DAFNe ResNet101 76.95 arXiv:2109.06148 PyTorch single stage, anchor free, center-to-corner regression
DAL+S2A-Net ResNet50 76.95 AAAI2021 PyTorch label assign ✅
GGHL DarkNet53 76.95 TIP PyTorch gaussian heatmap labeling ✅
EAutoDet DarkNet53 77.05 arXiv:2203.10747 - nas, yolov5 ✅
Yu et al. Res2Net50 77.18 JSTARS - boundary-aware vectors, centernet
CoF-Net ResNet50 77.2 TGRS - coarse-to-fine, geometric constraints, spatial-spectral nonocal features
DCL ResNet152 77.37 CVPR2021 TF boundary problem ✅
CSL+DCL ResNet152 77.37 IJCV TF boundary problem ✅
CLT-Det ResNet101 77.45 TGRS - transformer, correlation learning ✅
MSFF ResNet50 77.46 JSTARS - rotation invariance features
RIDet ResNet50 77.62 GRSL PyTorch, TF quad., representation ambiguity ✅
Oriented RepPoints Swin-Tiny 77.63 CVPR2022 PyTorch, MMRotate point-based, reppoints ✅
RSP ViTAEv2-S 77.72 arXiv:2204.02825 PyTorch remote sensing pretrain ✅
RDD ResNet101 77.75 Remote Sensing PyTorch rotation-decoupled
CenterOBB DLA-34 77.85 Remote Sensing - angle classification, centernet ✅
FSDet ResNet50 77.85 TGRS - label assign, feature refinemen, anchor free ✅
CG-Net ResNet101 77.89 arXiv:2103.11399 PyTorch attention
HSP ResNet101 78.01 TGRS - hierarchical semantic propagation
FoRDet VGG16 78.13 TGRS - refinenet
AProNet ResNet101 78.16 ISPRS Pyrotch axis projection-based angle learning, feature enhancement
MGAR DarkNet53 78.29 arXiv:2209.02884 - angle classification ✅
TransConvNet Swin Transformer 78.41 Remote Sensing - enhanced fpn, self attention, transformer
FR-Est ResNet101-DCN 78.49 TGRS - point-based estimator ✅
TIOE-Det - 78.69 ISPRS PyTorch misaligned tasks, orientation estimation ✅
DARDet ResNet50 78.74 GRSL PyTorch varifocalnet, dcn, piou
TS-Conv DarkNet53 78.75 arXiv:2209.02200 PyTorch dynamic label assignment, task-wise samping ✅
DHRec ResNet152 78.83 TPAMI PyTorch double horizontal rectangles
FCOSR ResNeXt101 78.90 arXiv:2111.10780 PyTorch anchor free
P2P ResNet101 79.15 AAAI2022 - poly iou loss ✅
SASM ResNeXt101 79.17 AAAI2022 MMRotate, PyTorch label assign ✅
AO2-DETR ResNet50 79.22 arXiv:2205.12785 PyTorch detr, feature refinement ✅
GSNet ResNet101 79.37 arXiv:2204.02033 - enhanced fpn
S2A-Net ResNet50/ResNet101 79.42/79.15 TGRS PyTorch, MMRotate refined single stage, feature alignment ✅
OFA-Net ResNet101 79.52 PRICAI2021 - refined single stage, feature alignment
TARDet ResNet101 79.57 CVPRW2022 - anchor free, feature alignment
RBox ResNet50 79.59 CVPR2022 - transformer, feature sampling and grouping ✅
O2DETR ResNet50 79.66 arXiv:2106.03146 - deformable detr, transformer ✅
ROSD ResNet101 79.76 Access - refined single stage, feature alignment
RAOD ResNet101 79.78 Applied Intelligence - enhanced fpn, feature refine
SES-Net ResNet50 79.80 arXiv:2111.03420 - sampling equivariance, self-attention ✅
SARA ResNet50/ResNet101 79.91/79.13 Remote Sensing - self-adaptive aspect ratio anchor, refine
ARP+R-EIoU YOLOv5x6 79.93 arXiv:2109.10187 - area ratio of parallelogram, R-EIoU, yolov5
GF-CSL ResNet101 79.94 TGRS PyTorch circular smooth label, gaussian focal loss ✅
ADT-Det ResNet152 79.95 Remote Sensing - feature pyramid transformer, feature refinement
ReDet ReR50-ReFPN 80.10 CVPR2021 PyTorch, MMRotate rotation-equivariant, rotation-invariant roI align ✅
G-Rep Swin-Tiny 80.12 arXiv:2205.11796 MMRotate pointset-based, gaussian modeling ✅
PCI ReR50-ReFPN 80.15 TIP - progressive context-dependent inference ✅
GWD ResNet152 80.23 ICML2021 TF, MMRotate boundary discontinuity, square-like problem, gaussian wasserstein distance loss ✅
DEA ReR50-ReFPN 80.37 TGRS PyTorch dynamic enhancement anchor ✅
FDOL ReR50-ReFPN 80.41 TGRS - frequency analysis, self-attention ✅
O2MER ResNet50 80.43 arXiv:2112.00504 - consistent geometric constraint
DODet ResNet50 80.62 TGRS PyTorch oriented proposal network, localization-guided head
KLD ResNet152 80.63 NeurIPS2021 TF, MMRotate Kullback-Leibler divergence, high-precision, scale invariance ✅
AOPG ResNet50/ResNet101 80.66/80.19 arXiv:2110.01931 PyTorch anchor free, feature align
Li et al. ResNet101 80.68 Remote Sensing - enhance fpn, polar
CGCDet ResNet50 80.70 TNNLS PyTorch geometric consistent constraint, label assignment ✅
AFF-Det ResNet50 80.73 Acm T Multim Comput. - enhanced fpn
PP-YOLOE-R-x CRN-x 80.73 arXiv:2211.02386 Paddle real-time, yolo ✅
Oriented R-CNN ResNet50/ResNet101 80.87/80.52 ICCV2021 PyTorch, MMRotate Rotation FPN, Gliding Vertex
OSKDet ResNet101 80.91 CVPR2022 - keypoint localization ✅
KFIoU Swin-Tiny 80.93 ICLR2023 TF, MMRotate Gaussian modeling, kalman filter ✅
QPDet ResNet50 81.00 TGRS PyTorch quadrant point regression, rotated box refinement ✅
Point RCNN ReR50-ReFPN/Swin-Tiny 80.71/81.32 Remote Sensing - point based, cascade rcnn ✅
SSEDet ResNet50 81.08 GRSL - circle theorem, short-side excursion ✅
RTMDet - 81.33 arXiv:2212.07784 MMRotate real-time ✅

DOTA1.0 (Task2)

Model Backbone mAP Paper Link Code Link Remark Recommend
FR-H (DOTA) ResNet101 60.46 CVPR2018 MXNet DOTA dataset, baseline ✅
Deep Active Learning ResNet18 64.26 arXiv:2003.08793 - CenterNet, Deep Active Learning ✅
SBL ResNet50 64.77 arXiv:1810.08103 - single stage
CenterFPANet ResNet18 65.29 HPCCT & BDAI 2020 arXiv:2009.03063 - light-weight
MARNet ResNet101 71.73 IJRS - based on scrdet
FMSSD VGG16 72.43 TGRS - IoU-based weighted loss, enhanced FPN
ICN ResNet101 72.45 ACCV2018 - image cascade, multi-scale ✅
IoU-Adaptive R-CNN ResNet101 72.72 Remote Sensing - IoU-based weighted loss, cascade
EFR VGG16 73.49 Remote Sensing PyTorch enhanced FPN
AF-EMS ResNet101 73.97 Remote Sensing - scale-aware feature, anchor free
SCRDet ResNet101 75.35 ICCV2019 TF attention, angular boundary problem ✅
FADet ResNet101 75.38 ICIP2019 - attention
MFIAR-Net ResNet152 76.07 Sensors - feature attention, enhanced FPN
F3-Net ResNet152 76.48 Remote Sensing - feature fusion and filtration
Mask OBB ResNeXt101 76.98 Remote Sensing - attention, multi-task ✅
CenterMap OBB ResNet101 77.33 TGRS - center-probability-map
ASSD VGG16 77.8 TGRS - feature aligned
AFC-Net ResNet101 78.06 Neurocomputing - adaptive feature concatenate
CG-Net ResNet101 78.26 arXiv:2103.11399 PyTorch attention
OPLD ResNet101 78.35 J-STARS PyTorch boundary problem, point-guided ✅
A2RMNet ResNet101 78.45 Remote Sensing - attention, enhanced FPN, different pooled sizes
OWSR Ensemble 78.79 CVPR2019 WorkShop - enhanced FPN
NPMMR-Det DarkNet53 78.93 TGRS PyTorch non-local, enhanced FPN
Parallel Cascade R-CNN ResNeXt101 78.96 Journal of Physics: Conference Series - cascade rcnn
DM-FPN ResNet-Based 79.27 Remote Sensing - enhanced FPN
DDBN ResNet101 79.3 PR - adjacent feature compensation, dual detection branch
SCRDet++ ResNet101 79.35 TPAMI TF denoising ✅
HSP ResNet101 80.42 TGRS - hierarchical semantic propagation
AFF-Det ResNet101 81.18 Acm T Multim Comput. - enhanced fpn

DOTA1.5 (Task1)

RBox-supervised Model Backbone mAP Paper Link Code Link Remark Recommend
OAN ResNet50 66.95 arXiv:2212.13136 PyTorch objectness activation network, efficient ✅
DAFNe ResNet101 71.99 arXiv:2109.06148 PyTorch r-fcos
RAOD ResNet101 74.70 Applied Intelligence - enhanced fpn, feature refine
OWSR Ensemble 76.60 CVPR2019 WorkShop - enhanced FPN
ReDet ReR50-ReFPN 76.80 CVPR2021 PyTorch rotation-equivariant, rotation-invariant RoI Align, ✅
FDOL ReR50-ReFPN 77.12 TGRS - frequency analysis, self-attention ✅
APE ResNeXt101(32x4) 78.34 TGRS arXiv:1906.09447 - length independent IoU (LIIoU) ✅
Point RCNN ReR50-ReFPN/Swin-Tiny 79.31/80.14 Remote Sensing - point based, cascade rcnn ✅

DOTA1.5 (Task2)

Model Backbone mAP Paper Link Code Link Remark Recommend
CDD-Net ResNet101 61.3 GRSL - attention
GCWNet ResNet101 62.7 GRSL - enhance fpn, feature refine
ReDet ReR50-ReFPN 78.08 CVPR2021 PyTorch rotation-equivariant, rotation-invariant RoI Align, ✅
OWSR Ensemble 79.50 CVPR2019 WorkShop - enhanced FPN

DOTA2.0 (Task1)

RBox-supervised Model Backbone mAP Paper Link Code Link Remark Recommend
OAN ResNet50 54.68 arXiv:2212.13136 PyTorch objectness activation network, efficient ✅
Wen et al. - 58.22 ICCECE 2022 - yolov5, csl

Related Articles

Model Paper Link Code Link Remark Recommend
SSSDET ICIP2019 arXiv:1909.00292 - vehicle detection, lightweight
AVDNet GRSL arXiv:1907.07477 - vehicle detection, small object
ClusDet ICCV2019 Caffe2 object cluster regions ✅
DMNet CVPR2020 WorkShop - object cluster regions ✅
AdaZoom arXiv:2106.10409 - object cluster regions, reinforcement learning ✅
OIS arXiv:1911.07732 related Pytorch code Oriented Instance Segmentation ✅
ISOP IGARSS2020 - Oriented Instance Segmentation
LR-RCNN arXiv:2005.14264 - vehicle detection -
GRS-Det TGRS - ship detection, rotation fcos -
DRBox arXiv:1711.09405 Caffe sar object detection ✅
DRBox-v2 TGRS TF sar object detection -
RAPiD arXiv:2005.11623 PyTorch overhead fisheye images -
OcSaFPN arXiv:2012.09859 - denoising -
CR2A-Net TGRS - ship detection -
- TGRS - knowledge distillation ✅
CHPDet arXiv:2101.11189 - new ship dataset ✅
EDRdet PRL - adaptive dynamic networks, coarse-to-fine ✅

Other Rotation Detection Codes and Benchmarks

Base Method Code Link
RetinaNet RetinaNet_Tensorflow_Rotation
YOLOv3 rotate-yolov3-Pytorch, YOLOv3-quadrangle-Pytorch, yolov3-polygon-Pytorch
YOLOv4 rotate-yolov4-Pytorch
YOLOv5 rotation-yolov5-Pytorch, YOLOv5_DOTA_OBB (CSL)
YOLOX YOLOX-Pytorch
YOLOv7 yolov7-obb (KLD)
CenterNet R-CenterNet-Pytorch
JDet Jittor
OBBDet_Swin OBBDet_Swin
OrientedRepPoints OrientedRepPoints_DOTA_Swin
Localization Distillation Rotated-LD

Dataset

Some remote sensing related object detection dataset statistics are in DATASET.md