Research Article
HRNet Encoder and Dual-Branch Decoder Framework-Based Scene Text Recognition Model
Table 1
The accuracy comparison between the proposed model and recent models (%).
| | Model | Benchmark | Average | | IIIT5k | SVT | IC03 | IC13 | IC15 | SVTP | CUTE80 | Regular | Irregular |
| | ASTER | 93.4 | 89.5 | 94.5 | 91.8 | 76.1 | 78.5 | 79.5 | 92.3 | 78.0 | | TextSR | 92.5 | 87.2 | 93.2 | 91.3 | 75.6 | 77.4 | 78.9 | 91.0 | 77.3 | | ESIR | 93.3 | 90.2 | — | 91.7 | 76.9 | 79.6 | 83.3 | 91.7 | 79.9 | | 2DOCR | 94 | 90.1 | 94.3 | 92.7 | 76.3 | 82.3 | 86.8 | 92.7 | 81.8 | | Bi-STET | 94.7 | 89 | 96 | 93.4 | 75.7 | 80.6 | 82.5 | 93.2 | 79.6 | | SEED | 93.8 | 89.6 | — | 92.8 | 80 | 81.4 | 83.6 | 92.0 | 81.6 | | DAN | 94.3 | 89.2 | 95 | 93.9 | 74.5 | 80 | 84.4 | 93.1 | 79.6 | | SPIN | 94.7 | 87.6 | 93.4 | 91.5 | 79.1 | 79.7 | 85.1 | 91.8 | 81.3 | | RobustScanner | 95.3 | 88.1 | — | 94.8 | 77.1 | 79.5 | 90.3 | 92.7 | 82.3 | | SCGAN | 94 | 90 | 95.6 | 93.3 | 81.6 | 85.1 | 78.1 | 93.2 | 81.6 | | Proposed model | 93.7 | 91.3 | 93.3 | 94.3 | 82.8 | 83.1 | 83.0 | 93.1 | 82.9 |
|
|
Note: bold font is the optimal value in each column, and the underline font is the suboptimal value in each column.
|
