Metraipolis › archive › dc_bc5c8eb15ccb

FlexReader: Towards Adaptive Decoding Paths for Efficient Arbitrary-Length Scene Text Recognition — arXiv:2610.41426

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.41426
License (as archived): CC0-1.0
Archived: 2026-11-03T05:46:22.762Z
SHA-256 of the archived text: 194115d91aae5ef3baf8fa3119b67bfebb6fc43b1fb5a3418a4930e36b00327f
Stable URL of this page: https://metraipolis.com/documento/dc_bc5c8eb15ccb

How to cite

FlexReader: Towards Adaptive Decoding Paths for Efficient Arbitrary-Length Scene Text Recognition — arXiv:2610.41426 — arXiv (en), https://arxiv.org/abs/2610.41426 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bc5c8eb15ccb (Metraipolis archive dc_bc5c8eb15ccb, sha256 194115d91aae5ef3baf8fa3119b67bfebb6fc43b1fb5a3418a4930e36b00327f)

BibTeX

@misc{dcbc5c8eb15ccb,
  title = {FlexReader: Towards Adaptive Decoding Paths for Efficient Arbitrary-Length Scene Text Recognition — arXiv:2610.41426},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41426},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bc5c8eb15ccb; sha256 194115d91aae5ef3baf8fa3119b67bfebb6fc43b1fb5a3418a4930e36b00327f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41426. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Existing Scene Text Recognition (STR) models are inherently coupled with a specific decoding strategy by their training paradigms, limiting their flexibility in deployment across diverse recognition scenarios. In this paper, we propose FlexReader, an STR model that supports multiple decoding strides, enabling a single model to achieve an optimal accuracy-efficiency trade-off across different datasets. Specifically, we first introduce a Sub-string Perception module to perceive the target sub-string from visual features, followed by multiple parallel decoding blocks that independently decode each character in the sub-string. During inference, FlexReader supports decoding with arbitrary strides within the perceived sub-string range. Furthermore, we propose Adaptive Prefix Decoding, which adaptively selects the decoding stride at each step based on the model's confidence in the decoded characters, thereby identifying the optimal decoding path for each input. Experimental results demonstrate that FlexReader exhibits robust recognition capability across a diverse range of benchmarks, including those featuring long text, curved text, and occlusions. It consistently outperforms existing decoding strategies in recognition accuracy while achieving up to 6-flod decoding speedup over autoregressive decoding.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.41426 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T05:33:32.294Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).