---
title: "A Visual Approach to Sketched Symbol Recognition"
authors:
  - "Tom Y. Ouyang"
  - "Randall Davis"
doi: ""
url: "https://ijcai.org/proceedings/2009/245"
pdf: "https://ijcai.org/proceedings/2009/0245.pdf"
venue: "Proceedings of the Twenty-First International Joint Conference on Artificial Intelligence"
year: 2009
pages: "1463-1467"
track: ""
keywords:
---

# A Visual Approach to Sketched Symbol Recognition

Tom Y. Ouyang, Randall Davis

*Proceedings of the Twenty-First International Joint Conference on Artificial Intelligence*, pages 1463-1467.

## Abstract

There is increasing interest in building systems that can automatically interpret hand-drawn sketches. However, many challenges remain in terms of recognition accuracy, robustness to different drawing styles, and ability to generalize across multiple domains. To address these challenges, we propose a new approach to sketched symbol recognition that focuses on the visual appearance of the symbols. This allows us to better handle the range of visual and stroke-level variations found in freehand drawings. We also present a new symbol classifier that is computationally efficient and invariant to rotation and local deformations. We show that our method exceeds state-of-the-art performance on all three domains we evaluated, including handwritten digits, PowerPoint shapes, and electrical circuit symbols.Tom Y. Ouyang, Randall Davis