List your company

Scene Text Recognition on SVTP (test)

98.3Word AccuracyCSD-B
81.03685.5189094.482Oct 10, 2019Sep 5, 2020Aug 2, 2021Jun 30, 2022May 27, 2023Apr 22, 2024Mar 20, 2025

Evaluation results

MethodLinks
CSD-BData=RBU, Params=104.9 M2025.0398.3
CLIP4STR-L*Training data=REBU-Syn2023.1298.13
CSD-DData=Real, Params=110.2 M2025.0398.1
CLIP4STR-LData=RBU, Params=446 M2025.0398.1
CSD-DData=RBU, Params=110.2 M2025.0398.1
CLIP4STR-BData=RBU, Params=158 M2025.0398
CLIP4STR-HData=RBU, Params=1 B2025.0398
CLIP4STR-B*Training data=REBU-Syn2023.1297.83
CSD-BData=Real, Params=104.9 M2025.0397.8
CSD-SData=RBU, Params=40.8 M2025.0397.8
CSD-SData=Real, Params=40.8 M2025.0397.5
CLIP4STR-LData=Real, Params=446 M2025.0397.4
CLIP4STR-LTraining data=Real2023.1297.36
CLIP4STR-BData=Real, Params=158 M2025.0397.2
CLIP4STR-BTraining data=Real2023.1296.74
PARSeqData=Real, Params=22.5 M2025.0395.7
PARSeqATraining data=Real, Reproduced=true2023.1294.73
MaskOCR(ViT-L)Training data=MJ+ST2023.1294.6
MAERec-BTraining data=Union14M-L2023.1294.4
ABINetTraining data=Real2023.1294.1
ABINETData=Real, Params=23.5 M2025.0393.9
TRBAData=Real, Params=49.6 M2025.0393.7
TrOCRLargeTraining data=MJ+ST+B2023.1293
DiG-ViT-BTraining data=Real2023.1292.9
ViTSTR-STraining data=Real2023.1291.5
VITSTR-SData=Real, Params=21.7 M2025.0391.4
MGP-STR FuseTraining Data=MJ and ST, Lexicon=None2022.0991.01
DiG-ViT-BTraining data=MJ+ST2023.1291
MATRN2021.1190.6
MATRNTraining data=MJ+ST2023.1290.6
MaskOCR(ViT-B)Training data=MJ+ST2023.1289.9
ABINet-LV+TPS++Training Data=90K+ST, Params=37.2M, Time=41.5ms2023.0589.6
MGP-STR VisionTraining Data=MJ and ST, Lexicon=None2022.0989.46
ABINetYear=20212021.1189.3
ABINet (reproduced)Implementation=reproduced2021.1189.3
ABINetTraining Data=MJ and ST, Lexicon=None2022.0989.3
ABINet-LVTraining Data=90K+ST, Params=36.7M, Time=37.2ms2023.0589.3
ABINetYear=2021, Training Data=90K+ST2021.1189.3
TRBATraining data=MJ+ST2023.1289.3
ABINetTraining data=MJ+ST2023.1289.3
ABINet + SeTaPruned %=28.1%2025.0389.3
NRTR+TPS++Training Data=90K+ST, Params=35.5M, Time=218ms2023.0589
PARSeqTraining Data=90K+ST, Params=23.8M, Time=11.8ms2023.0588.9
PARSeqATraining data=MJ+ST2023.1288.9
CDistNet w/o TPSTraining Data=90K+ST2021.1188.8
ABINet + InfoBatchPruned %=26.6%2025.0388.8
CDistNetTraining Data=90K+ST2021.1188.7
ABINetPruned %=0%2025.0388.7
ABINet + SeTaPruned %=71.0%2025.0388.2
ABINet + InfoBatchPruned %=50.3%2025.0388.1
ABINet + InfoBatchPruned %=38.1%2025.0388
ABINet + SeTaPruned %=40.4%2025.0388
GTRTraining Data=90K+ST, Params=42.1M, Time=18.8ms2023.0587.9
S-GTRYear=2022, Training Data=90K+ST2021.1187.9
PREN2DYear=20212021.1187.6
Pren2DTraining Data=90K+ST+Real, Time=67.4ms2023.0587.6
Pren2DYear=2021, Training Data=90K+ST+Real2021.1187.6
NRTRTraining Data=90K+ST, Params=31.7M, Time=212ms2023.0586.6
NRTRYear=2019, Training Data=90K+ST2021.1186.6
SATRNFeature map=2D, Training data=MJ+ST, Dictionary matching=False2019.1086.5
SATRNYear=20202021.1186.5
SATRNTraining Data=MJ and ST, Lexicon=None2022.0986.5
SARLexicon-free=true, Character-level annotations=false2021.0686.4
PETRTraining data=MJ+ST2023.1286.2
VisionLANYear=20212021.1186
VisionLANTraining Data=MJ and ST, Lexicon=None2022.0986
VisionLANTraining Data=90K+ST, Params=32.8M, Time=28.0ms2023.0586
VisionLANYear=2021, Training Data=90K+ST2021.1186
VisionLANTraining data=MJ+ST2023.1286
JVSRYear=20212021.1185.7
JVSRYear=2021, Training Data=90K+ST2021.1185.7
SRNBackbone=ResNet, Training Data=Synth90K+SynthText, Annotations=word-level, Lexicon=None2020.0385.1
SRNYear=20202021.1185.1
SRNTraining Data=MJ and ST, Lexicon=None2022.0985.1
SRNTraining Data=90K+ST, Params=49.3M, Time=26.9ms2023.0585.1
SRNYear=2020, Training Data=90K+ST2021.1185.1
SRNTraining data=MJ+ST2023.1285.1
TextScannerTraining data=MJ+ST2023.1284.8
MASTERLexicon-free=true, Character-level annotations=false2021.0684.5
SCATTERLexicon-free=true, Character-level annotations=false2021.0684.5
MASTERTraining Data=MJ and ST, Lexicon=None2022.0984.5
TextScannerTraining Data=MJ and ST, Lexicon=None2022.0984.3
TextscannerYear=2020, Training Data=90K+ST2021.1184.3
TPS-ResNet-BiLSTM-AttnTrain Dataset=S90k+ST+TextOCR2021.0584.19
STAR-NetTrain Dataset=S90k+ST+TextOCR2021.0583.88
SAR 3 x 20Lexicon-free=true, Character-level annotations=false2021.0683.7
Mask TextSpotterLexicon-free=true, Character-level annotations=true2021.0683.6
RCEEDLexicon-free=true, Character-level annotations=false2021.0683.6
GA-SPINYear=20212021.1183.2
SPINTraining Data=90K+ST2023.0583.2
SGBANetTraining Data=90K+ST2023.0583.1
RobustScannerLexicon-free=true, Character-level annotations=false2021.0682.9
Lyu et al. (Parallel)Backbone=ResNet, Training Data=Synth90K+SynthText, Annotations=word-level, Lexicon=None2020.0382.3
2D-AttentionYear=2019, Training Data=90K+ST2021.1182.3
Liao et al. (SAM)Backbone=ResNet, Training Data=Synth90K+SynthText, Annotations=word-level, Lexicon=None2020.0382.2
STANTraining Data=90K+ST2023.0582.2
MaskTextSpotterYear=2019, Training Data=90K+ST2021.1182.2
ViTSTRTraining Data=MJ and ST, Lexicon=None2022.0981.8
ViTSTR-BTraining data=MJ+ST2023.1281.8
AutoSTRYear=20202021.1181.7