IAM

Unknown

13,353 handwritten text lines from 657 writers. Standard handwriting benchmark.

Benchmark Stats

Models5
Papers8
Metrics2

SOTA History

wer

wer

Lower is better

RankModelSourceScoreYearPaper
1HTR-ConvText

HTR-ConvText. Dec 2024. Table 2 IAM test set.

Community12.92024Source
2HTR-VT

HTR-VT. Li et al. 2024. Table 4 IAM test set.

Community14.92024Source
3VAN

Vertical Attention Network (VAN). WER from comparison tables in HTR-VT (2409.08573) and HTR-ConvText (2512.05021). IAM line-level.

Community16.32022Source

cer

cer

Lower is better

RankModelSourceScoreYearPaper
1TrOCR-large

TrOCR-large (BEiT-large + RoBERTa-large). Microsoft. Table 4 in Li et al. 2023. IAM line-level test split. SOTA at publication.

Community2.892023Source
2TrOCR-base

TrOCR-base (BEiT-base + RoBERTa-base). Microsoft. Table 4 in Li et al. 2023. IAM line-level test split.

Community3.422023Source
3HTR-ConvText

HTR-ConvText (ConvText block hybrid CNN+Transformer). DAIR-Group, Dec 2024. Table 2 IAM test set. SOTA among no-pretraining methods.

Community42024Source
4HTR-VT

HTR-VT (Vision Transformer for HTR, no pre-training or synthetic data). Li et al. 2024. Table 4 IAM test set.

Community4.72024Source
5VAN

Vertical Attention Network (VAN). Coquenet et al. IEEE TPAMI 2022. IAM line-level CER from comparison tables in HTR-VT (2409.08573) and HTR-ConvText (2512.05021).

Community52022Source

Submit a Result