Knowledge Network Node

Texture and Shape Feature Fusion Based Sketch RecognitionChinese Full Text

ZHANG Xing-Yuan;HUANG Ya-Ping;ZOU Qi;PEI Yan-Ting;School of Computer and Information Technology, Beijing Jiaotong University;

Abstract: Human has a strong ability to recognize hand-drawn sketches. However, state-of-the-art models on sketch classification tasks remain challenging due to the sparse lines and limited details of sketches. Previous deep neural networks treat sketches as general images and ignore the shape representations for different categories. In this paper, we aim to address the problem by an end-to-end hand-drawn sketch recognition model, named dualmodel fusion network, which can capture both texture and shape information of sketches via a mutual learning strategy. Specifically, our model is composed of two branches: one branch can automatically extract texture features from an image-based representation, i.e., the raw sketches, and the other branch can obtain shape information from a graph-based representation, i.e., point-based sketches. Moreover, we propose an attention consistency loss to measure the attention heat-map consistency between the two branches, which can simultaneously enable the same concentration of discriminative regions in the two representations. Finally, the proposed dual-model fusion network is optimized by combining classification loss, category consistency loss and attention consistency loss. We conduct extensive experiments on two challenging data sets, TU-Berlin and Sketchy, for sketch classification tasks.Our dual-model fusion network significantly outperforms baselines, and achieves the new state-of-the-art performance.
  • DOI:

    10.16383/j.aas.c200070

  • Series:

    (I) Electronic Technology & Information Science

  • Subject:

    Computer Software and Application of Computer

  • Classification Code:

    TP391.41

  • Mobile Reading
    Read on your phone instantly
    Step 1

    Scan QR Codes

    "Mobile CNKI-CNKI Express" App

    Step 2

    Open“CNKI Express”

    and click the scan icon in the upper left corner of the homepage.

    Step 3

    Scan QR Codes

    Read this article on your phone.

  • CAJ Download
  • PDF Download

Download the mobile appuse the app to scan this coderead the article.

Tips: Please download CAJViewer to view CAJ format full text.

Download: 489 Page: 2223-2232 Pagecount: 10 Size: 1863K

Related Literature
  • Similar Article
  • Reader Recommendation
  • Associated Author