Out-of-Length Scene Text Recognition: A Two-Axis Diagnosis and a Training-Free Fix

• URL: http://arxiv.org/abs/2607.23194v1
• Date: Sat, 25 Jul 2026 13:16:57 GMT
• Title: Out-of-Length Scene Text Recognition: A Two-Axis Diagnosis and a Training-Free Fix
• Abstract summary: Scene Text Recognition models are trained almost exclusively on word crops of at most 25 characters.<n>This paper diagnoses that failure and then closes it.<n>We release the diagnosis harness and implementation.
• Score: 0.3437656066916039
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Scene Text Recognition (STR) models are trained almost exclusively on word crops of at most 25 characters, yet real deployments (signage, product labels, dense captions) require reading much longer text. This paper diagnoses that failure and then closes it. The diagnosis separates out-of-length failure into two simultaneously extrapolating axes (the encoder's width axis and the decoder's time axis) and shows that encoder width, not decoder length, is the dominant failure mode. Representation-side fixes bring only partial relief: training-free rotary rescalings recover at most 2-4 points of character error rate (CER), and a weighted fine-tuning recipe recovers 6-8 points while improving standard-benchmark accuracy, yet word accuracy on the Long Text Benchmark (LTB) stays near zero, because the residual gap lies in the decoding mechanism rather than the representation. We then close that gap at inference time, on an unmodified word-level checkpoint: the long image is sliced into overlapping crops at the model's training width, each decoded independently and in-distribution, and the reads stitched by geometry-anchored edit-distance alignment. This procedure reaches 42.79-43.05% bucket-average word accuracy on LTB across two base checkpoints, matching the published state of the art (41.57%) and beating it by 11-12 points on the hardest bucket, at wall-clock parity with plain decoding; applied unchanged to the public PARSeq checkpoint it reaches 47.11%. Once chunking is applied fine-tuning no longer helps: the decoding-side fix alone matches purpose-built architectures. We release the diagnosis harness and implementation.

Related papers

• Can Scene Text Recognition Read Rare Compositions? [1.7844382164707184]
  Scene text recognition is reported as 89--97% accurate on the six standard benchmarks, and the problem is widely treated as saturated.<n>We present an alternative reading.<n>When the same test images are stratified jointly by ground-truth word rarity and character n-gram novelty against a reference corpus, accuracy at the rareword x rare-trigram corner of the resulting 5x5 grid drops 10--18 pt below the q3/q3 centre.
  arXiv  Detail & Related papers  (2026-09-01T07:16:42Z)
• Confidence-Aware Ensemble and Long-Word Refinement for Artistic Text Recognition [0.6303112417588329]
  Artistic Text Recognition (ATR) remains challenging because word images often combine decorative fonts, curved layouts, object-like characters, clutter, and severe distortions.<n>This paper studies WordArt-V1.5 as a standardized benchmark for this setting and evaluates recent scene and artistic text recognizers under a common protocol.<n>We propose a confidence-aware ensemble that combines SVTRv2, PARSeq, and MAERec after fine-tuning on the official training split.
  arXiv  Detail & Related papers  (2026-08-30T18:56:04Z)
• Semantic Drift in Bug Resolution: How Behavioral Signals Propagate from Reports to Tests and Patches [18.727291516223115]
  Desc2Fix is a framework for measuring semantic alignment between bug reports, triggering tests, and developer-written fixes.<n>We analyze 2,857 report-test-patch triplets from Defects4J and SWT-Bench using GPT-4o and DeepSeek-Chat.
  arXiv  Detail & Related papers  (2026-07-20T22:32:23Z)
• FastOCR: Dynamic Visual Fixation via KV Cache Pruning for Efficient Document Parsing [51.905216364362325]
  We propose FastOCR, a training-free framework with two complementary modules.<n>FastOCR retains 98% of the unpruned model's accuracy while attending to only 5% of the visual tokens per decoding step.
  arXiv  Detail & Related papers  (2026-05-17T13:39:47Z)
• Enhancing IMU-Based Online Handwriting Recognition via Contrastive Learning with Zero Inference Overhead [4.519836503888727]
  We propose a training framework designed to improve feature representation and recognition accuracy without increasing inference costs.<n>ECHWR utilizes a temporary auxiliary branch that aligns sensor signals with semantic text embeddings during the training phase.<n> Evaluations on the OnHW-Words500 dataset show that ECHWR significantly outperforms state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-02-04T13:44:54Z)
• Context Perception Parallel Decoder for Scene Text Recognition [52.620841341333524]
  Scene text recognition methods have struggled to attain high accuracy and fast inference speed. We present an empirical study of AR decoding in STR, and discover that the AR decoder not only models linguistic context, but also provides guidance on visual context perception. We construct a series of CPPD models and also plug the proposed modules into existing STR decoders. Experiments on both English and Chinese benchmarks demonstrate that the CPPD models achieve highly competitive accuracy while running approximately 8x faster than their AR-based counterparts.
  arXiv  Detail & Related papers  (2023-07-23T09:04:13Z)
• SoftCorrect: Error Correction with Soft Detection for Automatic Speech Recognition [116.31926128970585]
  We propose SoftCorrect with a soft error detection mechanism to avoid the limitations of both explicit and implicit error detection. Compared with implicit error detection with CTC loss, SoftCorrect provides explicit signal about which words are incorrect. Experiments on AISHELL-1 and Aidatatang datasets show that SoftCorrect achieves 26.1% and 9.4% CER reduction respectively.
  arXiv  Detail & Related papers  (2022-12-02T09:11:32Z)
• A Comprehensive Handwritten Paragraph Text Recognition System: LexiconNet [3.9097549127191473]
  We have presented an efficient procedure using two state-of-the-art approaches from the literature of handwritten text recognition. The attention module is responsible for internal line segmentation that consequently processes a page in a line-by-line manner. At the decoding step, we have added a connectionist temporal classification-based word beam search decoder as a post-processing step.
  arXiv  Detail & Related papers  (2022-05-23T03:35:45Z)
• I2C2W: Image-to-Character-to-Word Transformers for Accurate Scene Text Recognition [68.95544645458882]
  This paper presents I2C2W, a novel scene text recognizer that is accurate and tolerant to various noises in scenes. I2C2W consists of an image-to-character module (I2C) and a character-to-word module (C2W) which are complementary and can be trained end-to-end.
  arXiv  Detail & Related papers  (2021-05-18T09:20:58Z)
• Rethinking Text Line Recognition Models [57.47147190119394]
  We consider two decoder families (Connectionist Temporal Classification and Transformer) and three encoder modules (Bidirectional LSTMs, Self-Attention, and GRCLs) We compare their accuracy and performance on widely used public datasets of scene and handwritten text. Unlike the more common Transformer-based models, this architecture can handle inputs of arbitrary length.
  arXiv  Detail & Related papers  (2021-04-15T21:43:13Z)
• Boosting Continuous Sign Language Recognition via Cross Modality Augmentation [135.30357113518127]
  Continuous sign language recognition deals with unaligned video-text pair. We propose a novel architecture with cross modality augmentation. The proposed framework can be easily extended to other existing CTC based continuous SLR architectures.
  arXiv  Detail & Related papers  (2020-10-11T15:07:50Z)
• Scene Text Recognition With Finer Grid Rectification [6.598317412802175]
  This paper proposed an end-to-end trainable model consists of a finer rectification module and a bidirectional attentional recognition network(Firbarn) The results of extensive evaluation on the standard benchmarks show Firbarn outperforms previous works, especially on irregular datasets.
  arXiv  Detail & Related papers  (2020-01-26T02:40:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.