List your company

Text-to-Image Retrieval on RSITMD

1mRRemoteCLIP
-1.092813.033627.1641.2864Dec 11, 2025Jan 2, 2026Jan 24, 2026Feb 15, 2026Mar 9, 2026Mar 31, 2026Apr 22, 2026

Evaluation results

MethodLinks
RemoteCLIPZero-shot protocol=True, Distribution setting=In-distribution2025.121———48.971
VLM2GeoVecZero-shot protocol=True, Distribution setting=Out-of-distribution2025.122———41.032.5
GeoRSCLIPZero-shot protocol=True, Distribution setting=Out-of-distribution2025.123———38.32.7
VLM2VecZero-shot protocol=True, Distribution setting=Out-of-distribution2025.124———34.624.2
CLIPZero-shot protocol=True, Distribution setting=Out-of-distribution2025.125———35.225
SkyCLIPZero-shot protocol=True, Distribution setting=Out-of-distribution2025.126———32.215.7
LW-MCRYear=2021, Pre-train data=02026.0413.553.8414.9625.73——
AMFMNYear=2022, Pre-train data=0, Lat. (ms)=150.1, Thr. (q/s)=6.72026.0415.514.9217.8328.33——
KAMCLYear=2023, Pre-train data=02026.0421.387.6822.3136.62——
CLIP-BaseYear=2021, Pre-train data=400M, large-scale VLM pre-training=true, Lat. (ms)=12.0, Thr. (q/s)=83.32026.0422.818.4627.1839.98——
CLIP-BaselineTraining Dataset=-, Zero-Shot=true2025.1223.860.08810.27850.4319——
VSE++Training Dataset=RSITMD, Zero-Shot=false2025.1224.540.07960.24870.3687——
LW-MCRTraining Dataset=RSITMD, Zero-Shot=false2025.1227.790.07790.30180.4978——
AMFMNTraining Dataset=RSITMD, Zero-Shot=false2025.1229.290.09960.34030.5296——
GaLRTraining Dataset=RSITMD, Zero-Shot=false2025.1231.410.11150.36680.5168——
GaLRYear=2022, Pre-train data=0, Lat. (ms)=291.7, Thr. (q/s)=3.42026.0431.4114.8231.6442.48——
SkyScriptYear=2024, Pre-train data=400M+2.6M, large-scale VLM pre-training=true, Lat. (ms)=13.5, Thr. (q/s)=74.12026.0431.4813.8133.1440.48——
HVSATraining Dataset=RSITMD, Zero-Shot=false2025.1233.150.11430.3920.5745——
SWANTraining Dataset=RSITMD, Zero-Shot=false2025.1233.890.11240.40490.606——
GeoRSCLIPTraining Dataset=RS5M, Zero-Shot=false2025.1235.680.14160.42390.5752——
FAAMIYear=2023, Pre-train data=02026.0435.9912.9642.3959.96——
VGSGNYear=2024, Pre-train data=0, Lat. (ms)=106.5, Thr. (q/s)=9.42026.0436.5113.2342.5763.43——
FAAMITraining Dataset=RSITMD, Zero-Shot=false2025.1236.520.12960.42390.5995——
PIRYear=2023, Pre-train data=0, Lat. (ms)=117.6, Thr. (q/s)=8.52026.0438.2412.1741.6863.41——
PIRTraining Dataset=RSITMD, Zero-Shot=false2025.1238.410.12770.41680.6341——
MTGFEYear=2023, Pre-train data=02026.0440.3817.1840.2753.48——
MTGFETraining Dataset=RSITMD, Zero-Shot=false2025.1240.780.16590.4850.6743——
MultilanguageTraining Dataset=RSITMD, Zero-Shot=false2025.1241.380.17610.49730.6659——
ML RetrievalTraining Dataset=RSITMD, Backbone=ViT-B/322026.0341.3817.6149.7360.59——
MSCMATYear=2024, Pre-train data=02026.0441.3815.1847.3564.73——
TRSLLaVATraining Dataset=-, Zero-Shot=true2025.1242.620.18260.43320.6841——
SIPNYear=2025, Pre-train data=0, Lat. (ms)=120.3, Thr. (q/s)=8.32026.0443.4717.0449.8766.95——
GRAFTYear=2024, Pre-train data=10.2M, large-scale VLM pre-training=true2026.0444.6124.9253.3961.47——
PETLYear=2023, Pre-train data=400M, large-scale VLM pre-training=true2026.0445.3320.4150.5368.54——
CLIPTraining Dataset=RET-3, Backbone=ViT-B/322026.0347.6322.6155.2769.87——
IERRYear=2024, Pre-train data=02026.0448.5524.4356.5972.66——
FTFYear=2026, Pre-train data=0, Candidate set size (K)=50, Lat. (ms)=24.7, Thr. (q/s)=40.62026.0448.6524.655.171.7——
RemoteCLIPTraining Dataset=RET-3 + DET-10 + SEG-4, Backbone=ViT-B/322026.0349.3822.1756.4667.41——
RemoteCLIPYear=2024, Pre-train data=400M+, large-scale VLM pre-training=true, Lat. (ms)=12.5, Thr. (q/s)=80.02026.0449.3822.1756.4673.41——
EAPAYear=2025, Pre-train data=400M, large-scale VLM pre-training=true2026.0449.9124.1657.4871.81——
GeoRSCLIPYear=2024, Pre-train data=400M+5M, large-scale VLM pre-training=true, Lat. (ms)=13.0, Thr. (q/s)=76.92026.0450.123.5457.5274.6——
FTFYear=2026, Pre-train data=0, Candidate set size (K)=100, Lat. (ms)=47.7, Thr. (q/s)=21.02026.0450.4926.757.872.34——
FSSNYear=2025, Pre-train data=0, Lat. (ms)=128.7, Thr. (q/s)=7.82026.0451.2825.7158.3275.4——
FTFYear=2026, Pre-train data=0, Candidate set size (K)=200, Lat. (ms)=93.7, Thr. (q/s)=10.72026.0451.4727.45973.52——
GeoRSCLIP-FTTraining Dataset=RSITMD + RSICD, Backbone=ViT-B/32, Evaluation Protocol=full fine-tuning2026.0351.8125.0457.8874.38——
AVIONTraining Dataset=RSITMD + RSICD, Backbone=ViT-B/32, Evaluation Protocol=lightweight prompt learning2026.0352.9226.0959.2275.41——
EBAKERYear=2024, Pre-train data=400M+0.2M, large-scale VLM pre-training=true, Lat. (ms)=13.6, Thr. (q/s)=73.92026.0453.3228.0560.3575.31——
Geo-RSClip+RS5Mzero-shot=true2025.10————38.02—
MaMMUT 400Mzero-shot=true, parameters=400M2025.10————32.7—
PIR-ITRzero-shot=true2025.10————39.85—
RS-MaMMut 400Mzero-shot=true, parameters=400M2025.10————42.58—
RS-SigLIP2 400Mzero-shot=true, parameters=400M2025.10————47.26—
SigLIP2 400Mzero-shot=true, parameters=400M2025.10————25.14—
SkyScript SkyCLIP-30zero-shot=true2025.10————30.58—