Text Recognition on Composite Dataset (IIIT, SVT, IC13, IC15, SVTP, CT)
98IIIT AccCCD-ViT-Small
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| CCD-ViT-SmallData=ARD, Params.=36M2022.11 | 98 | 96.4 | 98.3 | 90.3 | 92.7 | 98.3 | 95.57 | — | |
| CCD-ViT-BaseData=ARD, Params.=52M2022.11 | 98 | 97.8 | 98.3 | 91.6 | 96.1 | 98.3 | 96.3 | — | |
| CCD-ViT-SmallType=V (Language-free), Data=ARD, Params.=36M, Time (ms)=44.22022.11 | 98 | 96.4 | 98.3 | 90.3 | 92.7 | 98.3 | 97.27 | 95.13 | |
| CCD-ViT-BaseType=V (Language-free), Data=ARD, Params.=52M, Time (ms)=45.02022.11 | 98 | 97.8 | 98.3 | 91.6 | 96.1 | 98.3 | 97.83 | 95.99 | |
| DiG-ViT-SmallVenue=MM'22, Data=ARD, Params.=36M2022.11 | 97.7 | 96.1 | 97.3 | 88.6 | 91.6 | 96.2 | 94.69 | — | |
| DiG-ViT-BaseVenue=MM'22, Data=ARD, Params.=52M2022.11 | 97.6 | 96.5 | 97.6 | 88.9 | 92.9 | 96.5 | 94.92 | — | |
| CCD-ViT-BaseData=STD, Params.=52M2022.11 | 97.2 | 94.4 | 97 | 87.6 | 91.8 | 93.3 | 93.96 | — | |
| CCD-ViT-BaseType=V (Language-free), Data=STD, Params.=52M, Time (ms)=45.02022.11 | 97.2 | 94.4 | 97 | 87.6 | 91.8 | 93.3 | 96.02 | 93.48 | |
| CCD-ViT-TinyData=ARD, Params.=20M2022.11 | 97.1 | 96 | 97.5 | 87.5 | 91.6 | 95.8 | 94.18 | — | |
| CCD-ViT-TinyType=V (Language-free), Data=ARD, Params.=20M, Time (ms)=43.22022.11 | 97.1 | 96 | 97.5 | 87.5 | 91.6 | 95.8 | 96.34 | 93.65 | |
| PARSeqType=L (Language-aware), Venue=ECCV'22, Data=STD2022.11 | 97 | 93.6 | 96.2 | 86.5 | 88.9 | 92.2 | 95.28 | 92.65 | |
| CCD-ViT-SmallData=STD, Params.=36M2022.11 | 96.8 | 94.4 | 96.6 | 87.3 | 91.3 | 92.4 | 93.59 | — | |
| CCD-ViT-SmallType=V (Language-free), Data=STD, Params.=36M, Time (ms)=44.22022.11 | 96.8 | 94.4 | 96.6 | 87.3 | 91.3 | 92.4 | 95.63 | 93.11 | |
| DiG-ViT-SmallVenue=MM'22, Data=STD, Params.=36M2022.11 | 96.7 | 93.4 | 97.1 | 87.1 | 90.1 | 88.5 | 93.23 | — | |
| DiG-ViT-BaseVenue=MM'22, Data=STD, Params.=52M2022.11 | 96.7 | 94.6 | 96.9 | 87.1 | 91 | 91.3 | 93.49 | — | |
| SIGAType=V (Language-free), Venue=CVPR'23, Data=STD, Params.=113M, Time (ms)=56.32022.11 | 96.6 | 95.1 | 96.8 | 86.6 | 90.5 | 93.1 | 95.58 | 92.84 | |
| LevOCRType=L (Language-aware), Venue=ECCV'22, Data=STD, Params.=109M, Time (ms)=119.02022.11 | 96.6 | 92.9 | — | 86.4 | 88.1 | 91.7 | — | 92.26 | |
| CCD-ViT-TinyData=STD, Params.=20M2022.11 | 96.5 | 93.4 | 96.3 | 85.2 | 89.8 | 89.2 | 92.57 | — | |
| ABINet+ConCLRType=L (Language-aware), Venue=AAAI'22, Data=STD+WiKi2022.11 | 96.5 | 94.3 | — | 85.4 | 89.3 | 91.3 | — | 92.17 | |
| CCD-ViT-TinyType=V (Language-free), Data=STD, Params.=20M, Time (ms)=43.22022.11 | 96.5 | 93.4 | 96.3 | 85.2 | 89.8 | 89.2 | 94.96 | 91.98 | |
| DiG-ViT-TinyVenue=MM'22, Data=ARD, Params.=20M2022.11 | 96.4 | 94.4 | 96.2 | 87.4 | 90.2 | 94.1 | 93.37 | — | |
| MGP-STRType=V (Language-free), Venue=ECCV'22, Data=STD, Params.=148M, Time (ms)=12.32022.11 | 96.4 | 94.7 | — | 87.2 | 91 | 90.3 | — | 92.8 | |
| ABINetType=L (Language-aware), Venue=CVPR'21, Data=STD+WiKi, Params.=37M, Time (ms)=33.92022.11 | 96.2 | 93.5 | — | 86 | 89.3 | 89.2 | — | 92.02 | |
| Corner TransformerType=V (Language-free), Venue=ECCV'22, Data=STD, Params.=86M, Time (ms)=294.92022.11 | 95.9 | 94.6 | 96.4 | — | 91.5 | 92 | 95.13 | — | |
| DiG-ViT-TinyVenue=MM'22, Data=STD, Params.=20M2022.11 | 95.8 | 92.9 | 96.4 | 84.8 | 87.4 | 86.1 | 91.83 | — | |
| VisionLANType=L (Language-aware), Venue=ICCV'21, Data=STD, Params.=33M2022.11 | 95.8 | 91.7 | 95.7 | 83.7 | 86 | 88.5 | 93.8 | 90.64 | |
| S-GTRType=L (Language-aware), Venue=AAAI'22, Data=STD+WiKi, Params.=42M, Time (ms)=18.82022.11 | 95.8 | 94.1 | — | 84.6 | 87.9 | 92.3 | — | 91.5 | |
| PREN2DType=V (Language-free), Venue=CVPR'21, Data=STD, Time (ms)=67.42022.11 | 95.6 | 94 | — | 83 | 87.6 | 91.7 | — | 90.88 | |
| SGBANetType=V (Language-free), Venue=ECCV'22, Data=STD2022.11 | 95.4 | 89.1 | 95.1 | 83.1 | 88.2 | — | 92.83 | — | |
| PIMNetType=V (Language-free), Venue=MM'21, Data=STD, Time (ms)=28.42022.11 | 95.2 | 91.2 | 93.4 | 83.5 | 84.3 | 84.4 | 92.6 | 89.89 | |
| JVSRType=L (Language-aware), Venue=ICCV'21, Data=STD, Params.=44M, Time (ms)=26.32022.11 | 95.2 | 92.2 | 95.5 | 85.7 | 89.7 | — | 93.53 | — | |
| SRNType=L (Language-aware), Venue=CVPR'20, Data=STD, Params.=49M, Time (ms)=26.92022.11 | 94.8 | 91.5 | 95.5 | 82.7 | 85.1 | 87.8 | 93.07 | 89.74 | |
| Text is TextType=V (Language-free), Venue=ICCV'21, Data=STD2022.11 | 92.3 | 89.9 | — | — | 84.4 | 86.3 | — | — | |
| TRBAType=V (Language-free), Venue=CVPR'21, Data=STD, Params.=50M, Time (ms)=27.62022.11 | 92.1 | 88.9 | 93.1 | 78.3 | 79.5 | 78.2 | 89.74 | 85.97 | |
| PerSec-ViT*Venue=AAAI' 22, Data=STD, Pre-trained=Extra 100M images2022.11 | 88.1 | 86.8 | 94.2 | 73.6 | 77.7 | 72.7 | 83.77 | — | |
| SimANVenue=CVPR 22, Data=STD2022.11 | 87.5 | — | 89.9 | — | — | — | — | — | |
| SeqCLRVenue=CVPR'21, Data=STD2022.11 | 82.9 | — | 87.9 | — | — | — | — | — |