Comparative Analysis of Deep Learning-Based Models in Sketch Recognition
- DOI
- 10.2991/978-94-6463-512-6_34How to use a DOI?
- Keywords
- Graph Neural Networks; Sketch Recognition; Computer Vision
- Abstract
People usually use hand-drawn sketches to draw simple lines to express or record their ideas and intentions and create images or videos by hand-drawing. For some people who are not specialized in the field of art, the sketches of the same object will vary depending on the artistic style and drawing ability. As a result, they are often highly abstract, which makes the automatic recognition of sketches more difficult compared to other fields. Due to its visualization, sketch recognition has become an important hotspot problem, which effectively improves the efficiency and diversity of generation compared with the traditional manual creation method and becomes an important research direction within computer vision and graphics, and plays a crucial role in the fields of design and visual creation. Existing recognition approaches depend on hand-drawn features and depth features are deficient in recognizing their local information. By using a dataset QuickDraw, which consists of 345 categories with 50 million vector drawings released by Google, this work applies Graph Neural Networks (GNN) to improve the performance, improves the recognition accuracy of targeting sketches drawn by different people.
- Copyright
- © 2024 The Author(s)
- Open Access
- Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
Cite this article
TY - CONF AU - Fazhan Liu AU - Minxiao Lu AU - Wei Zhang PY - 2024 DA - 2024/09/23 TI - Comparative Analysis of Deep Learning-Based Models in Sketch Recognition BT - Proceedings of the 2024 International Conference on Artificial Intelligence and Communication (ICAIC 2024) PB - Atlantis Press SP - 310 EP - 322 SN - 1951-6851 UR - https://doi.org/10.2991/978-94-6463-512-6_34 DO - 10.2991/978-94-6463-512-6_34 ID - Liu2024 ER -