List your company

Text-to-Image Person Re-identification on CUHK-PEDES (test)

79.98Rank-1 Accuracy (R-1)BAMG
56.964862.939968.91574.8901May 15, 2023Oct 29, 2023Apr 14, 2024Sep 28, 2024Mar 15, 2025Aug 29, 2025Feb 13, 2026

Evaluation results

MethodLinks
BAMG2026.0179.9892.3194.03———
Our ICL*Training Sets=CUHK-PEDES2025.0577.9190.2794.1469.1353.4—
DiCo2026.0177.2191.8595.63———
RaSaImage Enc.=ALBEFF, Text Enc.=ALBEFF2023.0876.5190.2994.2569.38——
Our ICLTraining Sets=CUHK-PEDES2025.0576.4190.4894.3368.0451.99—
ICL2026.0176.4190.4894.33———
RDETraining Sets=CUHK-PEDES2025.0575.9490.1494.1267.5651.44—
RDE2026.0175.9490.1494.12———
RDENoise Rate=0%, Checkpoint Selection=Best2023.0875.9490.1494.1267.5651.44—
Our RDEImage Enc.=CLIP-ViT, Text Enc.=CLIP-Xformer2023.0875.9490.1494.1267.5651.44—
CFAM2026.0175.690.5394.36———
PLIPBackbone=Swin-B2023.0575.36—94.87———
PLOT2026.0175.2890.4294.12———
DCELImage Enc.=CLIP-ViT, Text Enc.=CLIP-Xformer2023.0875.0290.8994.52———
RDENoise Rate=20%, Checkpoint Selection=Last2023.0874.5389.2393.5566.1349.63—
RDENoise Rate=20%, Checkpoint Selection=Best2023.0874.4689.4293.6366.1349.66—
MUM2026.0174.2589.8393.58———
IRMmode=MTL, trained with multiple datasets=true2023.0674.2——66.5——
BiLMaImage Enc.=CLIP-ViT, Text Enc.=CLIP-Xformer2023.0874.0389.5993.6266.57——
PLIPBackbone=RN1522023.0573.68—94.22———
TBPS2026.0173.5488.1992.35———
TBPSImage Enc.=CLIP-ViT, Text Enc.=CLIP-Xformer2023.0873.5488.1992.3565.38——
IRRATraining Sets=CUHK-PEDES2025.0573.3889.9393.7166.1350.24—
IRRA2026.0173.3889.9393.71———
IRRANoise Rate=0%, Checkpoint Selection=Best2023.0873.3889.9393.7166.1350.24—
IRRAImage Enc.=CLIP-ViT, Text Enc.=CLIP-Xformer2023.0873.3889.9393.7166.1350.24—
RaSaTCLImage Enc.=TCL, Text Enc.=TCL2023.0873.2389.293.3266.43——
IRMmode=STL2023.0672.8——65.3——
PLIPBackbone=RN1012023.0572.34—93.46———
RDENoise Rate=50%, Checkpoint Selection=Best2023.0871.3387.4191.8163.547.36—
RDENoise Rate=50%, Checkpoint Selection=Last2023.0871.2587.3991.7663.5947.5—
PLIPBackbone=RN502023.0571.13—92.81———
DECLNoise Rate=20%, Checkpoint Selection=Best2023.0870.2987.0491.9362.8446.54—
DECLNoise Rate=20%, Checkpoint Selection=Last2023.0870.0887.292.1462.8646.63—
IRRANoise Rate=20%, Checkpoint Selection=Best2023.0869.7487.0992.262.2845.84—
CSKTVenue=CVPR20232026.0269.786.9291.862.74—47.68
RaSaimage shape=256x1282023.0669.6——63.9——
CFine2026.0169.5785.9391.15———
CFineNoise Rate=0%, Checkpoint Selection=Best2023.0869.5785.9391.15———
CFineImage Enc.=CLIP-ViT, Text Enc.=BERT2023.0869.5785.9391.15———
IRRANoise Rate=20%, Checkpoint Selection=Last2023.0869.4487.0992.0462.1645.7—
FlexiReIDVenue=ICML20252026.0269.286.4391.4162.4748.3247.86
UniReIDVenue=CVPR20232026.0268.7185.9390.84———
UniPTImage Enc.=CLIP-ViT, Text Enc.=CLIP-Xformer2023.0868.584.67————
LCR2SImage Enc.=RN50, Text Enc.=TextCNN2023.0867.3684.1989.6259.24——
CLIP-CNoise Rate=20%, Checkpoint Selection=Best2023.0866.4185.1590.8959.3643.02—
CLIP-CNoise Rate=20%, Checkpoint Selection=Last2023.0866.186.0191.0259.7743.57—
BEATImage Enc.=RN101, Text Enc.=BERT2023.0865.6183.4589.54———
IVT2026.0165.5983.1189.21———
LVTVenue=ECCV20222026.0265.5983.1189.21———
IVTNoise Rate=0%, Checkpoint Selection=Best2023.0865.5983.1189.21———
IVTImage Enc.=ViT-Base, Text Enc.=BERT2023.0865.5983.1189.21———
PBSLImage Enc.=RN50, Text Enc.=BERT2023.0865.3283.8189.26———
MLLMEmbed-ReIDQuery-Gallery feature type (C2C/E2E/E2C)=C2C2026.0265.2982.2287.5662.9252.1151.8
LGURVenue=MM20222026.0265.2583.1289———
LGURImage Enc.=DeiT-Small, Text Enc.=BERT2023.0865.2583.1289———
DECLNoise Rate=50%, Checkpoint Selection=Best2023.0865.2283.7289.2857.9441.39—
DECLNoise Rate=50%, Checkpoint Selection=Last2023.0865.0983.5889.2657.8941.35—
RDENoise=80%, Checkpoint Selection Strategy=Best2023.0864.9983.1589.5257.8441.07—
RDENoise=80%, Checkpoint Selection Strategy=Last2023.0864.9183.289.5457.8341.07—
AXM-NetImage Enc.=RN50, Text Enc.=BERT2023.0864.4480.5286.7758.73——
CAIBCVenue=MM20222026.0264.4382.8787.35———
CAIBCImage Enc.=RN50, Text Enc.=BERT2023.0864.4382.8788.37———
TIPCB2026.0164.2683.1989.1———
TIPCBVenue=NeurIPS20222026.0264.2683.1989.1———
TIPCBImage Enc.=RN50, Text Enc.=BERT2023.0864.2683.1989.1———
SAF2026.0164.1382.6288.4———
SAFVenue=ICASSP20222026.0264.1382.6288.458.61—42.73
SAFImage Enc.=ViT-Base, Text Enc.=BERT2023.0864.1382.6288.4———
SAF2023.0664.1——58.6——
PSLD2023.0664.1——60.1——
Han et al.Image Enc.=CLIP-RN101, Text Enc.=CLIP-Xformer2023.0864.0881.7388.1960.08——
SRCF2026.0164.0482.9988.81———
LBUL+BERTVenue=MM20222026.0264.0482.6687.22———
LBULImage Enc.=RN50, Text Enc.=BERT2023.0864.0482.6687.22———
CLIP-CNoise Rate=50%, Checkpoint Selection=Best2023.0864.0283.6689.3857.3340.9—
CLIP-CNoise Rate=50%, Checkpoint Selection=Last2023.0863.9783.7489.5457.3540.88—
ISANetImage Enc.=RN50, Text Enc.=LSTM2023.0863.9282.1587.69———
LapscoreImage Enc.=RN50, Text Enc.=BERT2023.0863.4—87.8———
BLIPBackbone=ViT-L2023.0563.23—88.74———
IRRANoise Rate=50%, Checkpoint Selection=Best2023.0862.4182.2388.455.5238.48—
SSAN2026.0161.3780.1586.73———
SSANNoise Rate=0%, Checkpoint Selection=Best2023.0861.3780.1586.73———
SSANImage Enc.=RN50, Text Enc.=LSTM2023.0861.3780.1586.73———
BLIPBackbone=ViT-B2023.0560.37—86.63———
ALBEF2023.0660.3——56.7——
DSSLVenue=MM20212026.0259.9880.4187.56———
DSSLImage Enc.=RN50, Text Enc.=BERT2023.0859.9880.4187.56———
NAFS2026.0159.9479.8686.7———
LUPBackbone=RN1522023.0559.41—86.22———
NAFSImage Enc.=RN50, Text Enc.=BERT2023.0859.3679.138654.07——
LUP-NLBackbone=RN1522023.0559.27—86.06———
CLIPBackbone=ViT-B2023.0558.91—86.21———
MLLMEmbed-ReIDQuery-Gallery feature type (C2C/E2E/E2C)=E2C2026.0258.877.3584.9858.0347.8747.55
BaselineBackbone=RN1522023.0558.64—85.84———
IVTNoise Rate=20%, Checkpoint Selection=Best2023.0858.5978.5185.6157.1945.78—
LUP-NLBackbone=RN1012023.0558.06—85.64———
CLIPBackbone=RN1012023.0558.04—85.88———
CLIPBackbone=RN502023.0557.92—85.12———
SOLIDERBackbone=Swin-B2023.0557.85—84.97———