Zero-Shot Chinese Character Recognition with Stroke-Level Decomposition
Authors: Jingye Chen, Bin Li, Xiangyang Xue
IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the proposed method on handwritten characters, printed artistic characters, and scene characters. The experimental results validate that the proposed method outperforms existing methods on both character zero-shot and radical zero-shot tasks. |
| Researcher Affiliation | Academia | Jingye Chen, Bin Li , Xiangyang Xue Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science, Fudan University {jingyechen19, libin, xyxue}@fudan.edu.cn |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks labeled as such. |
| Open Source Code | No | The paper does not provide any concrete access information, such as a repository link or explicit statement, regarding the release of its source code. |
| Open Datasets | Yes | HWDB1.0-1.1 [Liu et al., 2013] contains 2,678,424 offline handwritten Chinese character images. ICDAR2013 [Yin et al., 2013] contains 224,419 offline handwritten Chinese character images. CTW [Yuan et al., 2019] contains Chinese characters collected from street views. |
| Dataset Splits | Yes | From HWDB1.0-1.1, we choose samples with labels in the first m classes of 3,755 characters as the training set, where m ranges in {500,1000,1500,2000,2755}. From ICDAR2013, we choose samples with labels in the last 1000 classes as the test set. |
| Hardware Specification | Yes | We implement our method with Py Torch and conduct experiments on an NVIDIA RTX 2080Ti GPU with 11GB memory. |
| Software Dependencies | No | The paper mentions "Py Torch" but does not provide specific version numbers for it or any other software dependencies. |
| Experiment Setup | Yes | The Adadelta optimizer is used with the learning rate set to 1. The batch size is set to 32. Each input image is resized to 32 32 and normalized to [-1,1]. We adopt a weight decay rate of 10 4 in zero-shot settings to avoid overfitting. |