List your company

Image-to-text Retrieval on RSITMD

1RankRemoteCLIP
-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-distribution2025.12143.211———
VLM2GeoVecZero-shot protocol=True, Distribution setting=Out-of-distribution2025.12231.862.5———
GeoRSCLIPZero-shot protocol=True, Distribution setting=Out-of-distribution2025.12335.692.7———
VLM2VecZero-shot protocol=True, Distribution setting=Out-of-distribution2025.12429.274.2———
CLIPZero-shot protocol=True, Distribution setting=Out-of-distribution2025.12526.335———
SkyCLIPZero-shot protocol=True, Distribution setting=Out-of-distribution2025.12626.035.7———
LW-MCRYear=2021, Pre-train data=02026.0413.55——4.3313.2519.18
AMFMNYear=2022, Pre-train data=0, Lat. (ms)=150.1, Thr. (q/s)=6.72026.0415.51——4.9814.1720.75
KAMCLYear=2023, Pre-train data=02026.0421.38——6.4218.6936.58
CLIP-BaseYear=2021, Pre-train data=400M, large-scale VLM pre-training=true, Lat. (ms)=12.0, Thr. (q/s)=83.32026.0422.81——9.1120.4231.7
CLIP-BaselineTraining Dataset=-, Zero-Shot=true2025.1223.86——9.5321.0332.74
VSE++Training Dataset=RSITMD, Zero-Shot=false2025.1224.54——10.2827.6539.6
LW-MCRTraining Dataset=RSITMD, Zero-Shot=false2025.1227.79——10.1828.9839.82
AMFMNTraining Dataset=RSITMD, Zero-Shot=false2025.1229.29——11.0629.0238.72
GaLRTraining Dataset=RSITMD, Zero-Shot=false2025.1231.41——14.8231.6442.48
GaLRYear=2022, Pre-train data=0, Lat. (ms)=291.7, Thr. (q/s)=3.42026.0431.41——11.1536.6851.68
SkyScriptYear=2024, Pre-train data=400M+2.6M, large-scale VLM pre-training=true, Lat. (ms)=13.5, Thr. (q/s)=74.12026.0431.48——12.0535.7453.68
HVSATraining Dataset=RSITMD, Zero-Shot=false2025.1233.15——13.232.0845.58
SWANTraining Dataset=RSITMD, Zero-Shot=false2025.1233.89——13.3532.5545.11
GeoRSCLIPTraining Dataset=RS5M, Zero-Shot=false2025.1235.68——19.0334.5146.46
FAAMIYear=2023, Pre-train data=02026.0435.99——16.1535.6248.89
VGSGNYear=2024, Pre-train data=0, Lat. (ms)=106.5, Thr. (q/s)=9.42026.0436.51——14.1634.9650.66
FAAMITraining Dataset=RSITMD, Zero-Shot=false2025.1236.52——19.3235.6248.89
PIRYear=2023, Pre-train data=0, Lat. (ms)=117.6, Thr. (q/s)=8.52026.0438.24——18.1441.1552.88
PIRTraining Dataset=RSITMD, Zero-Shot=false2025.1238.41——17.6441.1553.82
MTGFEYear=2023, Pre-train data=02026.0440.38——16.3348.1566.85
MTGFETraining Dataset=RSITMD, Zero-Shot=false2025.1240.78——17.9240.9353.32
MultilanguageTraining Dataset=RSITMD, Zero-Shot=false2025.1241.38——19.6940.2654.42
ML RetrievalTraining Dataset=RSITMD, Backbone=ViT-B/322026.0341.38——19.0940.2654.42
MSCMATYear=2024, Pre-train data=02026.0441.38——22.3542.9255.75
TRSLLaVATraining Dataset=-, Zero-Shot=true2025.1242.62——19.5444.9161.28
SIPNYear=2025, Pre-train data=0, Lat. (ms)=120.3, Thr. (q/s)=8.32026.0443.47——23.0146.0257.96
GRAFTYear=2024, Pre-train data=10.2M, large-scale VLM pre-training=true2026.0444.61——25.2544.6857.89
PETLYear=2023, Pre-train data=400M, large-scale VLM pre-training=true2026.0445.33——24.1647.1261.28
CLIPTraining Dataset=RET-3, Backbone=ViT-B/322026.0347.63——23.675063.27
IERRYear=2024, Pre-train data=02026.0448.55——27.6648.8961.06
FTFYear=2026, Pre-train data=0, Candidate set size (K)=50, Lat. (ms)=24.7, Thr. (q/s)=40.62026.0448.65——27.849.363.4
RemoteCLIPTraining Dataset=RET-3 + DET-10 + SEG-4, Backbone=ViT-B/322026.0349.38——27.8850.6665.71
RemoteCLIPYear=2024, Pre-train data=400M+, large-scale VLM pre-training=true, Lat. (ms)=12.5, Thr. (q/s)=80.02026.0449.38——27.8850.6665.71
EAPAYear=2025, Pre-train data=400M, large-scale VLM pre-training=true2026.0449.91——29.8753.3262.83
GeoRSCLIPYear=2024, Pre-train data=400M+5M, large-scale VLM pre-training=true, Lat. (ms)=13.0, Thr. (q/s)=76.92026.0450.1——30.0951.5563.27
FTFYear=2026, Pre-train data=0, Candidate set size (K)=100, Lat. (ms)=47.7, Thr. (q/s)=21.02026.0450.49——29.651.265.3
FSSNYear=2025, Pre-train data=0, Lat. (ms)=128.7, Thr. (q/s)=7.82026.0451.28——30.5352.2165.49
FTFYear=2026, Pre-train data=0, Candidate set size (K)=200, Lat. (ms)=93.7, Thr. (q/s)=10.72026.0451.47——30.352.466.2
GeoRSCLIP-FTTraining Dataset=RSITMD + RSICD, Backbone=ViT-B/32, Evaluation Protocol=full fine-tuning2026.0351.81——32.353.3267.92
AVIONTraining Dataset=RSITMD + RSICD, Backbone=ViT-B/32, Evaluation Protocol=lightweight prompt learning2026.0352.92——33.1754.4669.18
EBAKERYear=2024, Pre-train data=400M+0.2M, large-scale VLM pre-training=true, Lat. (ms)=13.6, Thr. (q/s)=73.92026.0453.32——34.0754.267.95
Geo-RSClip+RS5Mzero-shot=true2025.10—33.33————
MaMMUT 400Mzero-shot=true, parameters=400M2025.10—28.83————
PIR-ITRzero-shot=true2025.10—38.64————
RS-MaMMut 400Mzero-shot=true, parameters=400M2025.10—42.63————
RS-SigLIP2 400Mzero-shot=true, parameters=400M2025.10—43.14————
SigLIP2 400Mzero-shot=true, parameters=400M2025.10—24.62————
SkyScript SkyCLIP-30zero-shot=true2025.10—30.75————