{
  "id": "2204.05535",
  "version": "v1",
  "published": "2022-04-12T05:43:46.000Z",
  "updated": "2022-04-12T05:43:46.000Z",
  "title": "Open-set Text Recognition via Character-Context Decoupling",
  "authors": [
    "Chang Liu",
    "Chun Yang",
    "Xu-Cheng Yin"
  ],
  "comment": "Accepted at CVPR 2022 (Poster)",
  "categories": [
    "cs.CV"
  ],
  "abstract": "The open-set text recognition task is an emerging challenge that requires an extra capability to cognize novel characters during evaluation. We argue that a major cause of the limited performance for current methods is the confounding effect of contextual information over the visual information of individual characters. Under open-set scenarios, the intractable bias in contextual information can be passed down to visual information, consequently impairing the classification performance. In this paper, a Character-Context Decoupling framework is proposed to alleviate this problem by separating contextual information and character-visual information. Contextual information can be decomposed into temporal information and linguistic information. Here, temporal information that models character order and word length is isolated with a detached temporal attention module. Linguistic information that models n-gram and other linguistic statistics is separated with a decoupled context anchor mechanism. A variety of quantitative and qualitative experiments show that our method achieves promising performance on open-set, zero-shot, and close-set text recognition datasets.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2022-04-12T05:43:46.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "character-context decoupling",
      "contextual information",
      "close-set text recognition datasets",
      "visual information",
      "open-set text recognition task"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}