List your company
ResearchFormerly WizWand

Scene Classification on RESISC-45 (TR=20%)

96.36AccuracySelectiveMAE
88.50890.546592.58594.6235Mar 18, 2024Jul 26, 2024Dec 4, 2024Apr 14, 2025Aug 22, 2025Dec 31, 2025May 11, 2026
Updated 5mo ago

Evaluation results

MethodLinks
SelectiveMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=533k2024.0696.36
LSKNet-S#P=14.4M, FLOPS=54.4G2024.0395.83
SelectiveMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=533k, Pre-training=OpticalRS-13 (4M)2024.0695.77
LSKNet-T#P=4.3M, FLOPS=19.2G2024.0395.75
RVSABackbone=VIT-B+RVSA [21], Params (M)=862024.0695.69
RingMoBackbone=Swin-B [102], Params (M)=882024.0695.67
RVSA#P=114.4M, FLOPS=301.3G2024.0395.66
RSP-VITAE#P=19.3M, FLOPS=119.1G2024.0395.6
MBENet#P=23.9M, FLOPS=108.5G2024.0395.58
SelectiveMAEBackbone=ViT-B [2], Params (M)=86, Throughput/Minute=556k2024.0695.48
CTNet2024.0395.4
FENet#P=23.9M, FLOPS=92.0G2024.0395.39
KFBNet2024.0395.11
ScaleMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=206k2024.0695.04
RSP-R50#P=25.6M, FLOPS=86.3G2024.0395.02
F2BRBM#P=25.6M, FLOPS=86.3G2024.0394.87
ConvNext#P=28.0M, FLOPS=93.7G2024.0394.76
FSCNet#P=28.8M, FLOPS=166.1G2024.0394.76
GRMANet#P=54.1M, FLOPS=171.4G2024.0394.72
SatLasBackbone=Swin-B [102], Params (M)=88, Throughput/Minute=243k2024.0694.7
MBLANet2024.0394.66
GFMBackbone=Swin-B [102], Params (M)=882024.0694.64
SelectiveMAEBackbone=ViT-B [2], Params (M)=86, Throughput/Minute=556k, Pre-training=OpticalRS-13 (4M)2024.0694.58
RSP-Swin#P=27.5M, FLOPS=37.7G2024.0394.51
MAEBackbone=ViT-B [2], Params (M)=86, Throughput/Minute=264k, Pre-training=OpticalRS-13 (4M)2024.0694.43
EAM#P=>42.3M, FLOPS=>164.32024.0394.29
SatMAEBackbone=ViT-L [2], Params (M)=307, Throughput/Minute=205k2024.0694.1
ViT-B#P=86.0M, FLOPS=118.9G2024.0393.96
TOVBackbone=ResNet-50 [101], Params (M)=262024.0693.79
UPetu#P=87.7M, FLOPS=>322.2G2024.0393.79
IDCCP#P=25.6M, FLOPS=86.3G2024.0393.76
MSANet#P=>42.3M, FLOPS=>164.32024.0393.52
LSENet#P=25.9M, FLOPS=>86.3G2024.0393.34
RingMo-LiteGeneral pre-training=-, Remote sensing pre-training=0.15M, Parameters=30M (Swin-T)2026.0593.25
MA-FE#P=>25.6M, FLOPS=>86.3G2024.0393.21
GASSLBackbone=ResNet-50 [101], Params (M)=262024.0693.06
MG-CAP#P=>42.3M, FLOPS=>164.3G2024.0392.95
SeCoBackbone=ResNet-50 [101], Params (M)=262024.0692.91
SCCov#P=13.0M2024.0392.1
CACOBackbone=ResNet-50 [101], Params (M)=262024.0691.94
SSL4EOBackbone=ViT-S [2], Params (M)=222024.0691.27
Foundation ModelGeneral pre-training=-, Remote sensing pre-training=0.28M, Parameters=32M (Swin-T)2026.0588.81