# Optical Character Recognition on Vertical Japanese prose

Source: [Optical Character Recognition on Vertical Japanese prose](<https://www.sota2.com/research/sota/optical-character-recognition-on-vertical-japanese-prose>)

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

Data updated: 2026-02-26T05:39:06.381Z

Evaluation of Optical Character Recognition performance on the Vertical Japanese prose dataset, measured by Character Error Rate.

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 | Optical Character Recognition |
| Dataset (including any stated split) | Vertical Japanese prose |
| Metric | Character Error Rate |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 6 of 6 |
| 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

- [Character Error Rate](<https://www.sota2.com/research/sota/optical-character-recognition-on-vertical-japanese-prose.md?metric=Character%20Error%20Rate>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Character Error Rate | Result date | Report paper |
| --- | --- | --- | --- |
| EffOCR-T (Base) | 2.7 | Not recorded | 68fa74401e209131cea60891 |
| EffOCR-C (Base) | 3 | Not recorded | 68fa74401e209131cea60891 |
| EffOCR-C (Small) | 3.6 | Not recorded | 68fa74401e209131cea60891 |
| Google Cloud Vision OCR | 13.5 | Not recorded | 68fa74401e209131cea60891 |
| Baidu OCR | 17.7 | Not recorded | 68fa74401e209131cea60891 |
| Tesseract OCR (Best) | 74.4 | Not recorded | 68fa74401e209131cea60891 |

### EffOCR-T (Base)

Character Error Rate: 2.7. Result ID: 699337683a6c314849147c3d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

Recorded evaluation configuration: {"Seq2Seq?":"X","Transformer?":"✓","Pretraining":"from scratch","Parameters":"101.8 M"}. This record does not assert independent reproduction.

### EffOCR-C (Base)

Character Error Rate: 3. Result ID: 699337683a6c314849147c3b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

Recorded evaluation configuration: {"Seq2Seq?":"X","Transformer?":"X","Pretraining":"from scratch","Parameters":"112.5 M"}. This record does not assert independent reproduction.

### EffOCR-C (Small)

Character Error Rate: 3.6. Result ID: 699337683a6c314849147c3c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

Recorded evaluation configuration: {"Seq2Seq?":"X","Transformer?":"X","Pretraining":"from scratch","Parameters":"9.3 M"}. This record does not assert independent reproduction.

### Google Cloud Vision OCR

Character Error Rate: 13.5. Result ID: 699337683a6c314849147c3e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

Recorded evaluation configuration: {"Pretraining":"off-the-shelf"}. This record does not assert independent reproduction.

### Baidu OCR

Character Error Rate: 17.7. Result ID: 699337683a6c314849147c3f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

Recorded evaluation configuration: {"Pretraining":"off-the-shelf"}. This record does not assert independent reproduction.

### Tesseract OCR (Best)

Character Error Rate: 74.4. Result ID: 699337683a6c314849147c40.

[Reporting paper](<https://www.sota2.com/research/paper/68fa74401e209131cea60891.md>)

Recorded evaluation configuration: {"Pretraining":"off-the-shelf","Parameters":"1.4 M"}. This record does not assert independent reproduction.

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