List your company
ResearchFormerly WizWand

Scene Classification on AID TR=50%

98.6AccuracyFENet
61.648871.241980.83590.4281May 6, 2022Sep 11, 2022Jan 18, 2023May 27, 2023Oct 3, 2023Feb 9, 2024Jun 17, 2024
Updated 8mo ago

Evaluation results

MethodLinks
FENet#P=23.9M, FLOPS=92.0G2024.0398.6
MGML-FENetPretrain=IMP, Backbone=DenseNet-121, Method=MGML-FENet [77]2022.0898.6
ViT-BPretrain=MAE, Backbone=ViT-B, Method=—2022.0898.56
MBENet#P=23.9M, FLOPS=108.5G2024.0398.54
SelectiveMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=533k2024.0698.52
RVSABackbone=VIT-B+RVSA [21], Params (M)=862024.0698.5
SelectiveMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=533k, Pre-training=OpticalRS-13 (4M)2024.0698.5
RVSA#P=114.4M, FLOPS=301.3G2024.0398.5
ViTAE-B + RVSA♦Pretrain=MAE, Backbone=ViTAE-B + RVSA*, Method=—2022.0898.5
ViTAE-B + RVSAPretrain=MAE, Backbone=ViTAE-B + RVSA, Method=—2022.0898.48
ViT-B + RVSA♦Pretrain=MAE, Backbone=ViT-B + RVSA*, Method=—2022.0898.44
ViTAE-BPretrain=MAE, Backbone=ViTAE-B, Method=—2022.0898.42
RingMoPretrain=RingMo [32], Backbone=ViT-B, Method=—2022.0898.38
RingMoBackbone=Swin-B [102], Params (M)=882024.0698.34
RingMoPretrain=RingMo [32], Backbone=Swin-B, Method=—2022.0898.34
ViTAE-B + VSAPretrain=MAE, Backbone=ViTAE-B + VSA, Method=—2022.0898.34
ViT-B + RVSAPretrain=MAE, Backbone=ViT-B + RVSA, Method=—2022.0898.33
RSP-Swin#P=27.5M, FLOPS=37.7G2024.0398.3
ViT-B + VSAPretrain=MAE, Backbone=ViT-B + VSA, Method=—2022.0898.3
SelectiveMAEBackbone=ViT-B [2], Params (M)=86, Throughput/Minute=556k2024.0698.28
RSP-VITAE#P=19.3M, FLOPS=119.1G2024.0398.22
LSKNet-S#P=14.4M, FLOPS=54.4G2024.0398.22
VITAEv2-SPretrain=RSP, Backbone=VITAEv2-S, Method=—2022.0898.22
LSKNet-T#P=4.3M, FLOPS=19.2G2024.0398.14
SelectiveMAEBackbone=ViT-B [2], Params (M)=86, Throughput/Minute=556k, Pre-training=OpticalRS-13 (4M)2024.0698.12
Swin-TPretrain=IMP, Backbone=Swin-T, Method=—2022.0898.1
VITAEv2-SPretrain=IMP, Backbone=VITAEv2-S, Method=—2022.0898.08
MAEBackbone=ViT-B [2], Params (M)=86, Throughput/Minute=264k, Pre-training=OpticalRS-13 (4M)2024.0698.02
RSP-R50#P=25.6M, FLOPS=86.3G2024.0397.89
CTNet2024.0397.7
ScaleMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=206k2024.0697.58
FSCNet#P=28.8M, FLOPS=166.1G2024.0397.51
AGOSType=CNN, Backbone=DenseNet-1212022.0597.43
KFBNet2024.0397.4
ConvNext#P=28.0M, FLOPS=93.7G2024.0397.4
GRMANet#P=54.1M, FLOPS=171.4G2024.0397.39
GRMANetPretrain=IMP, Backbone=ResNet-50, Method=GRMANet [73]2022.0897.39
SatLasBackbone=Swin-B [102], Params (M)=88, Throughput/Minute=243k2024.0697.38
AGOSType=CNN, Backbone=ResNet-1012022.0597.22
CADNetType=CNN, Year=20202022.0597.16
MBLANet2024.0397.14
MBLANetType=CNN, Year=20212022.0597.14
TOVBackbone=ResNet-50 [101], Params (M)=262024.0697.09
GFMBackbone=Swin-B [102], Params (M)=882024.0697.09
EAM#P=>42.3M, FLOPS=>164.32024.0397.06
UPetu#P=87.7M, FLOPS=>322.2G2024.0397.06
EAMPretrain=IMP, Backbone=ResNet-101, Method=EAM [74]2022.0897.06
AGOSType=CNN, Backbone=ResNet-502022.0597.01
F2BRBM#P=25.6M, FLOPS=86.3G2024.0396.97
F2BRBMPretrain=IMP, Backbone=ResNet-50, Method=F2BRBM [5]2022.0896.97
IDCCP#P=25.6M, FLOPS=86.3G2024.0396.95
SatMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=205k2024.0696.94
MSDFFType=CNN, Year=20202022.0596.74
Contourlet CNNType=CNN, Year=20212022.0596.65
LSENet#P=25.9M, FLOPS=>86.3G2024.0396.36
LSENetPretrain=IMP, Backbone=VGG-16, Method=LSENet [42]2022.0896.36
LiGNetType=CNN, Year=20212022.0596.19
MG-CAP#P=>42.3M, FLOPS=>164.3G2024.0396.12
MG-CAPType=CNN, Year=20202022.0596.12
SCCov#P=13.0M2024.0396.1
ViT-B#P=86.0M, FLOPS=118.9G2024.0396.08
ViT-BPretrain=IMP, Backbone=ViT-B, Method=—2022.0896.08
MSANet#P=>42.3M, FLOPS=>164.32024.0396.01
MSANetPretrain=IMP, Backbone=ResNet-101, Method=MSANet [75]2022.0896.01
SeCoBackbone=ResNet-50 [101], Params (M)=262024.0695.99
MA-FE#P=>25.6M, FLOPS=>86.3G2024.0395.98
MF2NetType=CNN, Year=20202022.0595.93
GASSLBackbone=ResNet-50 [101], Params (M)=262024.0695.92
LSENetType=CNN, Year=20212022.0595.82
DMSMILType=CNN, Year=20212022.0595.65
CSPTPretrain=CSPT [80], Backbone=ViT-L, Method=—2022.0895.62
GBNetType=CNN, Year=20202022.0595.48
CSPTPretrain=CSPT [80], Backbone=ViT-B, Method=—2022.0895.11
CACOBackbone=ResNet-50 [101], Params (M)=262024.0695.05
MS2APType=CNN, Year=20212022.0594.82
SSL4EOBackbone=ViT-S [2], Params (M)=222024.0694.74
DSENetType=CNN, Year=20212022.0594.5
MSCPType=CNN, Year=20182022.0594.42
ASPPretrain=ASP [76], Backbone=ResNet-101, Method=—2022.0894.2
RBFFPretrain=IMP, Backbone=MobileNet-V2 [78], Method=RBFF [79]2022.0893.64
ARCNetType=RNN, Year=20182022.0593.1
MIDCNetType=CNN, Year=20202022.0592.53
RANetType=CNN, Year=20202022.0592.35
APNetType=CNN, Year=20192022.0592.15
SPPNetType=CNN, Year=2017, Note=Conducted by authors2022.0591.45
TEXNetType=CNN, Year=20172022.0590
VGGNet-16Type=CNN, Year=20172022.0589.64
AlexNetType=CNN, Year=20172022.0589.53
GoogLeNetType=CNN, Year=20172022.0586.39
AGANType=GAN, Year=20192022.0584.52
MARTAType=GAN, Year=20172022.0581.57
LDA(SIFT)Type=Hand-crafted, Year=20172022.0568.96
BOVW(SIFT)Type=Hand-crafted, Year=20172022.0568.37
PLSA(SIFT)Type=Hand-crafted, Year=20172022.0563.07