# Scene Text Recognition on SVT Cleaned (test)

Source: [Scene Text Recognition on SVT Cleaned (test)](<https://www.sota2.com/research/sota/scene-text-recognition-on-svt-cleaned-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:09:00.058Z

Evaluation of Scene Text Recognition performance on the test split of the SVT Cleaned dataset, measured by Word Accuracy.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Scene Text Recognition |
| Dataset (including any stated split) | SVT Cleaned (test) |
| Metric | Word Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Word Accuracy](<https://www.sota2.com/research/sota/scene-text-recognition-on-svt-cleaned-test.md?metric=Word%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Word Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| MPSTR | 98.5 | Not recorded | 68fa765f1e209131cea6cb50 |
| ABINet | 97.8 | Not recorded | 68fa765f1e209131cea6cb50 |
| CLIP4STR-B | 97.8 | Not recorded | 68fa765f1e209131cea6cb50 |
| CLIP4STR-L | 97.8 | Not recorded | 68fa765f1e209131cea6cb50 |
| PARSeq | 97.5 | Not recorded | 68fa765f1e209131cea6cb50 |

### MPSTR

Word Accuracy: 98.5. Result ID: 69941c703a6c3148491b221a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77b61e209131cea74eba.md>)

Recorded evaluation configuration: {"Training Data":"3.3M real samples"}. This record does not assert independent reproduction.

### ABINet

Word Accuracy: 97.8. Result ID: 69941c703a6c3148491b2218.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa697f1e209131cea227bf.md>)

[Implementation](<https://github.com/FangShancheng/ABINet>)

Recorded evaluation configuration: {"Training Data":"3.3M real samples"}. This record does not assert independent reproduction.

### CLIP4STR-B

Word Accuracy: 97.8. Result ID: 69941c703a6c3148491b221b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb50.md>)

[Implementation](<https://github.com/VamosC/CLIP4STR>)

Recorded evaluation configuration: {"Training Data":"3.3M real samples","Backbone":"ViT-B"}. This record does not assert independent reproduction.

### CLIP4STR-L

Word Accuracy: 97.8. Result ID: 69941c703a6c3148491b221c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb50.md>)

[Implementation](<https://github.com/VamosC/CLIP4STR>)

Recorded evaluation configuration: {"Training Data":"3.3M real samples","Backbone":"ViT-L"}. This record does not assert independent reproduction.

### PARSeq

Word Accuracy: 97.5. Result ID: 69941c703a6c3148491b2219.

[Reporting paper](<https://www.sota2.com/research/paper/68fa765f1e209131cea6cb50.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d031e209131cea37551.md>)

[Implementation](<https://github.com/baudm/parseq>)

Recorded evaluation configuration: {"Training Data":"3.3M real samples"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.