Abstract
This paper studies the challenging problem of zero-shot Chinese text recognition, which requires the model to train on text line images containing only the seen characters, and then recognize the unseen characters from new text line images. Most of the previous methods only consider the zero-shot Chinese character recognition problem. They attempt to decompose the Chinese characters into radical representations and then recognize them at the radical level. Some methods developed recently have extended the radical-based recognition model from recognizing characters to recognizing text lines. However, the disadvantages of these methods include the requirement of long training time and a complicated decoding process. In addition, these methods are unsuitable for long text sequences. In this paper, we have proposed a novel zero-shot Chinese text recognition network (ZCTRN) by matching the class embeddings with the visual features. Specifically, our proposed model consists of three components: a text line encoder that extracts the visual features from the text line images, a class embedding module that encodes the character classes into class embeddings, and a bidirectional embedding transfer module that can map the class embeddings into the visual space and preserve the information of the original class embeddings. In addition, we use a distance-based CTC decoder to match the visual features with the class embeddings and output the recognition results. Experimental obtained by applying our proposed network to the MTHv2 dataset and the ICDAR-2013 handwriting competition dataset show that our method not only preserves high accuracy in recognizing text line images containing seen characters, but also outperforms the existing state-of-the-art models in recognizing text line images containing unseen characters.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Akata, Z., Perronnin, F., Harchaoui, Z., Schmid, C.: Label-embedding for attribute-based classification. In: CVPR, pp. 819–826 (2013)
Ao, X., Zhang, X., Yang, H., Yin, F., Liu, C.: Cross-modal prototype learning for zero-shot handwriting recognition. In: ICDAR, pp. 589–594 (2019)
Baek, J., et al.: What is wrong with scene text recognition model comparisons? Dataset and model analysis. In: ICCV, pp. 4714–4722 (2019)
Bahdanau, D., Cho, K., Bengio, Y.: Neural machine translation by jointly learning to align and translate. In: ICLR (2015)
Cao, Z., Lu, J., Cui, S., Zhang, C.: Zero-shot handwritten Chinese character recognition with hierarchical decomposition embedding. Pattern Recognit. 107, 107488 (2020)
Du, J., Wang, Z.-R., Zhai, J., Hu, J.: Deep neural network based hidden Markov model for offline handwritten Chinese text recognition. In: ICPR, pp. 3428–3433 (2016)
Fu, Y., Hospedales, T.M., Xiang, T., Gong, S.: Transductive multi-view zero-shot learning. IEEE Trans. Pattern Anal. Mach. Intell. 37(11), 2332–2345 (2015)
Graves, A., Fernández, S., Gomez, F., Schmidhuber, J.: Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks. In: ICML, pp. 369–376 (2006)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR, pp. 770–778 (2016)
Kodirov, E., Xiang, T., Gong, S.: Semantic autoencoder for zero-shot learning. In: CVPR, pp. 3174–3183 (2017)
Lampert, C.H., Nickisch, H., Harmeling, S.: Learning to detect unseen object classes by between-class attribute transfer. In: CVPR, pp. 951–958. IEEE (2009)
Li, Z., Wu, Q., Xiao, Y., Jin, M., Lu, H.: Deep matching network for handwritten Chinese character recognition. Pattern Recognit. 107, 107471 (2020)
Liu, C., Yin, F., Wang, D., Wang, Q.: CASIA online and offline Chinese handwriting databases. In: ICDAR, pp. 37–41 (2011)
Ma, W., Zhang, H., Jin, L., Wu, S., Wang, J., Wang, Y.: Joint layout analysis, character detection and recognition for historical document digitization. In: ICFHR, pp. 31–36 (2020)
Messina, R., Louradour, J.: Segmentation-free handwritten Chinese text recognition with LSTM-RNN. In: ICDAR, pp. 171–175 (2015)
Peng, D., Jin, L., Wu, Y., Wang, Z., Cai, M.: A fast and accurate fully convolutional network for end-to-end handwritten Chinese text segmentation and recognition. In: ICDAR, pp. 25–30. IEEE (2019)
Shi, B., Bai, X., Yao, C.: An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition. IEEE Trans. Pattern Anal. Mach. Intell. 39(11), 2298–2304 (2017)
Shi, B., Yang, M., Wang, X., Lyu, P., Yao, C., Bai, X.: ASTER: an attentional scene text recognizer with flexible rectification. IEEE Trans. Pattern Anal. Mach. Intell. 41(9), 2035–2048 (2018)
Shigeto, Y., Suzuki, I., Hara, K., Shimbo, M., Matsumoto, Y.: Ridge regression, hubness, and zero-shot learning. In: Appice, A., Rodrigues, P.P., Santos Costa, V., Soares, C., Gama, J., Jorge, A. (eds.) ECML PKDD 2015, Part I. LNCS (LNAI), vol. 9284, pp. 135–151. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-23528-8_9
Wan, Z., Xie, F., Liu, Y., Bai, X., Yao, C.: 2D-CTC for scene text recognition. arXiv preprint arXiv:1907.09705 (2019)
Wang, Q., Yin, F., Liu, C.: Handwritten Chinese text recognition by integrating multiple contexts. IEEE Trans. Pattern Anal. Mach. Intell. 34(8), 1469–1481 (2012)
Wang, S., Chen, L., Xu, L., Fan, W., Sun, J., Naoi, S.: Deep knowledge training and heterogeneous CNN for handwritten Chinese text recognition. In: ICFHR, pp. 84–89. IEEE (2016)
Wang, T., Xie, Z., Li, Z., Jin, L., Chen, X.: Radical aggregation network for few-shot offline handwritten Chinese character recognition. Pattern Recognit. Lett. 125, 821–827 (2019)
Wang, T., et al.: Decoupled attention network for text recognition. In: AAAI, pp. 12216–12224 (2020)
Wang, W., Zheng, V.W., Yu, H., Miao, C.: A survey of zero-shot learning: settings, methods, and applications. ACM Trans. Intell. Syst. Technol. 10(2), 1–37 (2019)
Wang, Z.R., Du, J., Wang, J.M.: Writer-aware CNN for parsimonious hmm-based offline handwritten Chinese text recognition. Pattern Recognit. 100, 107102 (2020)
Wang, Z.R., Du, J., Wang, W.C., Zhai, J.F., Hu, J.S.: A comprehensive study of hybrid neural network hidden Markov model for offline handwritten Chinese text recognition. Int. J. Doc. Anal. Recognit. 21(4), 241–251 (2018)
Wu, Y.C., Yin, F., Chen, Z., Liu, C.L.: Handwritten Chinese text recognition using separable multi-dimensional recurrent neural network. In: ICDAR, vol. 1, pp. 79–84. IEEE (2017)
Xie, C., Lai, S., Liao, Q., Jin, L.: High performance offline handwritten Chinese text recognition with a new data preprocessing and augmentation pipeline. In: Bai, X., Karatzas, D., Lopresti, D. (eds.) DAS 2020. LNCS, vol. 12116, pp. 45–59. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-57058-3_4
Xie, Z., Huang, Y., Zhu, Y., Jin, L., Liu, Y., Xie, L.: Aggregation cross-entropy for sequence recognition. In: CVPR, pp. 6531–6540 (2019)
Xiu, Y., Wang, Q., Zhan, H., Lan, M., Lu, Y.: A handwritten Chinese text recognizer applying multi-level multimodal fusion network. In: ICDAR, pp. 1464–1469 (2019)
Yin, F., Wang, Q.F., Zhang, X.Y., Liu, C.L.: ICDAR 2013 Chinese handwriting recognition competition. In: ICDAR, pp. 1464–1470. IEEE (2013)
Zeiler, M.D.: Adadelta: an adaptive learning rate method. arXiv preprint arXiv:1212.5701 (2012)
Zhang, J., Du, J., Dai, L.: Radical analysis network for learning hierarchies of Chinese characters. Pattern Recognit. 103, 107305 (2020)
Zhang, J., Zhu, Y., Du, J., Dai, L.: Radical analysis network for zero-shot learning in printed Chinese character recognition. In: ICME, pp. 1–6. IEEE (2018)
Zhang, L., Xiang, T., Gong, S.: Learning a deep embedding model for zero-shot learning. In: CVPR, pp. 2021–2030 (2017)
Acknowledgement
This research is supported in part by NSFC (Grant No.: 61936003), the National Key Research and Development Program of China (No. 2016YFB1001405), GD-NSF (no. 2017A030312006), Guangdong Intellectual Property Office Project (2018-10-1).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Huang, Y., Jin, L., Peng, D. (2021). Zero-Shot Chinese Text Recognition via Matching Class Embedding. In: Lladós, J., Lopresti, D., Uchida, S. (eds) Document Analysis and Recognition – ICDAR 2021. ICDAR 2021. Lecture Notes in Computer Science(), vol 12823. Springer, Cham. https://doi.org/10.1007/978-3-030-86334-0_9
Download citation
DOI: https://doi.org/10.1007/978-3-030-86334-0_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-86333-3
Online ISBN: 978-3-030-86334-0
eBook Packages: Computer ScienceComputer Science (R0)