Scene Text Recognition on SVTP 645 (test)
97.67AccuracyPARSeq-H
Evaluation results
| Method | Links | ||
|---|---|---|---|
| PARSeq-HParameters=0.6B, Train Data=REBU-Syn2023.12 | 97.67 | — | |
| PARSeq-LTrain Data=REBU-Syn2023.12 | 97.21 | — | |
| PARSeq-BTrain Data=REBU-Syn2023.12 | 96.74 | — | |
| PARSeq-SParameters=22M, Train Data=REBU-Syn2023.12 | 95.97 | — | |
| PARSeq_ATrain data=R2022.07 | 95.7 | — | |
| PARSeq_NTrain data=R2022.07 | 94.6 | — | |
| ABINetTrain data=R2022.07 | 93.9 | — | |
| TRBATrain data=R2022.07 | 93.7 | — | |
| ABINet++ (LV)Labeled Datasets=MJ+ST+Real, Vision Model Scale=LV2022.11 | 92.2 | — | |
| VITSTR-STrain data=R2022.07 | 91.4 | — | |
| ABINet++† (LV)Labeled Datasets=MJ+ST, Unlabeled Datasets=Uber-Text, Vision Model Scale=LV, Training Strategy=self-training2022.11 | 90.1 | — | |
| ABINet++‡ (LV)Labeled Datasets=MJ+ST, Unlabeled Datasets=Uber-Text, Vision Model Scale=LV, Training Strategy=ensemble self-training2022.11 | 89.9 | — | |
| ABINetTrain data=S32022.07 | 89.3 | — | |
| ABINet++ (LV)Labeled Datasets=MJ+ST, Vision Model Scale=LV2022.11 | 89.3 | 33.9 | |
| CVAE-FeedTrain data=S2022.07 | 88.9 | — | |
| PARSeq_ATrain data=S2022.07 | 88.9 | — | |
| PARSeq_NTrain data=S2022.07 | 87.9 | — | |
| SRN* (LV)Labeled Datasets=MJ+ST, Reproduced=true, Vision Model Scale=LV2022.11 | 87.9 | 26.9 | |
| PREN2DTrain data=S2022.07 | 87.6 | — | |
| ABINetTrain data=S2022.07 | 87.1 | — | |
| ABINet++ (SV)Labeled Datasets=MJ+ST, Vision Model Scale=SV2022.11 | 87 | 31.6 | |
| TRBATrain data=S2022.07 | 86.9 | — | |
| SATRNTraining Data=MJ+ST, Annotations=word2021.02 | 86.5 | — | |
| SRN* (SV)Labeled Datasets=MJ+ST, Reproduced=true, Vision Model Scale=SV2022.11 | 86.4 | 24.2 | |
| STN-CSTRTraining Data=MJ+ST, Annotations=word, STN=true2021.02 | 86.2 | — | |
| STN-CSTRTrain data=S2022.07 | 86.2 | — | |
| VisionLANTrain data=S2022.07 | 86 | — | |
| VisionLan Wang et al.Year=2021, Labeled Datasets=MJ+ST2022.11 | 86 | 11.5 | |
| PIMNet Qiao et al.Year=2021, Labeled Datasets=MJ+ST+Real2022.11 | 85.9 | — | |
| Bhunia et al.Train data=S2022.07 | 85.7 | — | |
| Bhunia et al.Year=2021, Labeled Datasets=MJ+ST2022.11 | 85.7 | — | |
| GTC Hu et al.Year=2020, Labeled Datasets=MJ+ST+Real2022.11 | 85.7 | — | |
| SRNTraining Data=MJ+ST, Annotations=word2021.02 | 85.1 | — | |
| SRNTrain data=S2022.07 | 85.1 | — | |
| SRN Yu et al.Year=2020, Labeled Datasets=MJ+ST2022.11 | 85.1 | 46.2 | |
| TextScannerTrain data=S*2022.07 | 84.8 | — | |
| Textscanner Wan et al.Year=2020, Labeled Datasets=MJ+ST+Real2022.11 | 84.8 | — | |
| PlugNetTrain data=S2022.07 | 84.3 | — | |
| Textscanner Wan et al.Year=2020, Labeled Datasets=MJ+ST2022.11 | 84.3 | — | |
| PIMNet Qiao et al.Year=2021, Labeled Datasets=MJ+ST2022.11 | 84.3 | 19.1 | |
| VITSTR-STrain data=S2022.07 | 83.9 | — | |
| RCEEDTrain data=S,B2022.07 | 83.6 | — | |
| Robust ScannerTrain data=S,B2022.07 | 82.9 | — | |
| VITSTR-BTrain data=S22022.07 | 81.8 | — | |
| AutoSTRTrain data=S2022.07 | 81.7 | — | |
| SEEDTraining Data=MJ+ST, Annotations=word, Beam search=true2021.02 | 81.4 | — | |
| SEED Qiao et al.Year=2020, Labeled Datasets=MJ+ST2022.11 | 81.4 | 53.3 | |
| CRNNTrain data=R2022.07 | 80.6 | — | |
| DAN Wang et al.Year=2020, Labeled Datasets=MJ+ST2022.11 | 80 | — | |
| TRBATrain data=S2022.07 | 79.5 | — | |
| RobustScanner Yue et al.Year=2020, Labeled Datasets=MJ+ST2022.11 | 79.5 | 36 | |
| Baek et al.Training Data=MJ+ST, Annotations=word2021.02 | 79.2 | — | |
| CRNNTrain data=S2022.07 | 73.5 | — | |
| CRNNTrain data=S2022.07 | 64.8 | — |