Spike-HTR: Spiking Neural Transformer for Handwritten Text Recognition
2026-08-03 • Neural and Evolutionary Computing
Neural and Evolutionary Computing
AI summaryⓘ
The authors address how recognizing handwritten text can be tricky because most image areas are blank, and many parts of the sequence contain no information. They introduce Spike-HTR, a new system that turns static images of handwriting into a series of simpler steps for a special type of neural network called a spiking neural network. Their method focuses only on important parts of the handwriting to reduce work and gets good accuracy without needing extra help like language models. This approach was tested on several datasets and showed promising results.
Handwritten Text RecognitionSpiking Neural NetworksSequence Length ReductionCTC (Connectionist Temporal Classification)Static Image ProcessingDeep Sequence MixerInkCoderCharacter Error Rate (CER)Blank-Dominated SequencesHybrid Neural Networks
Authors
Xiubo Liang, Jinxing Han, Yuke Li, Haoqi Zhu, Yu Zhao, Hongzhi Wang
Abstract
Handwritten Text Recognition (HTR) is computationally imbalanced in two ways: most image pixels are background, and many width-axis sequence positions are blank-dominated. This creates a mismatch for Spiking Neural Networks (SNNs): handwriting is observed as a static image, whereas spiking computation unfolds over timesteps. We propose Spike-HTR, a hybrid spiking recognizer that controls both the number of spiking steps and the number of width positions processed by the deep sequence mixer. To make a static image suitable for short-horizon spiking inference, InkCoder converts it into a coarse-to-fine input stream, where early steps cover broad stroke regions and later steps emphasize sharper stroke details. To reduce sequence computation, a CTC-guided length reducer keeps likely character or uncertain positions and compresses long blank-dominated stretches before deep mixing. With $T{=}2$, Spike-HTR trains only on target data, decodes without language models or lexicons, and reaches validation/test CERs of 3.5/5.4, 2.3/2.5, and 4.2/3.9 on IAM, LAM, and READ2016. Codes are available at https://github.com/QomolangmaH/SpikeHTR.