Sketch recognition

About: Sketch recognition is a research topic. Over the lifetime, 1611 publications have been published within this topic receiving 40284 citations.

Training-Free Synthesized Face Sketch Recognition Using Image Quality Assessment Metrics.

Abstract: Face sketch synthesis has wide applications ranging from digital entertainments to law enforcements. Objective image quality assessment scores and face recognition accuracy are two mainly used tools to evaluate the synthesis performance. In this paper, we proposed a synthesized face sketch recognition framework based on full-reference image quality assessment metrics. Synthesized sketches generated from four state-of-the-art methods are utilized to test the performance of the proposed recognition framework. For the image quality assessment metrics, we employed the classical structured similarity index metric and other three prevalent metrics: visual information fidelity, feature similarity index metric and gradient magnitude similarity deviation. Extensive experiments compared with baseline methods illustrate the effectiveness of the proposed synthesized face sketch recognition framework. Data and implementation code in this paper are available online at www.ihitworld.com/WNN/IQA_Sketch.zip.

Mechanix: The Development of a Sketch Recognition Truss Tutoring System

Abstract: Mechanix is a sketch recognition tool that provides an efficient means for engineering students to learn how to draw truss free-body diagrams (FBDs) and solve truss problems. The system allows for students to sketch these FBDs, as they normally would by hand, into a tablet computer; a mouse can also be used for regular computer monitors. Mechanix is able to provide immediate and intelligent feedback to the students; it tells them if they are missing any components of the FBD. The program is also able to tell students whether their solved reaction forces or member forces are correct or not without actually providing the answers. Mechanix also has a checklist feature which appears in the same window as the program, it guides the students through the problem and automatically updates as the student progresses and solves each part of the truss problem. This paper presents a study to evaluate the effectiveness and advantages of using Mechanix in the classroom as a supplement to traditional teaching and learning methods. Freshman engineering classes were recruited for this experiment and were divided into an experimental group (students who used Mechanix in class and for their assignments) and a control group (students who were not exposed to Mechanix). The learning gains between these two groups were evaluated using a series of quantitative formal assessments which include concept inventories and homework, quiz, and exam grades. Qualitative data was also collected through focus groups for both groups to gather the students’ impressions of the programs for the experimental group and general teaching styles for the control group. Due to some issues with the server that runs Mechanix, the students were not able to properly use Mechanix during the in-class evaluations. We believe that this caused the results to show that there was no change in the homework and concept inventory scores between both groups for the current evaluation. However, the results show that Mechanix is a capable tool for enhancing students learning and performance in exams. The focus group discussion showed that the students really liked the program; they mostly appreciated the instant feedback and they said that Mechanix motivated them to move on to more problems when they saw that they had successfully solved the previous ones.

DPFrag: Trainable Stroke Fragmentation Based on Dynamic Programming

Abstract: Many computer graphics applications must fragment freehand curves into sets of prespecified geometric primitives. For example, sketch recognition typically converts hand-drawn strokes into line and arc segments and then combines these primitives into meaningful symbols for recognizing drawings. However, current fragmentation methods' shortcomings make them impractical. For example, they require manual tuning, require excessive computational resources, or produce suboptimal solutions that rely on local decisions. DPFrag is an efficient, globally optimal fragmentation method that learns segmentation parameters from data and produces fragmentations by combining primitive recognizers in a dynamic-programming framework. The fragmentation is fast and doesn't require laborious and tedious parameter tuning. In experiments, it beat state-of-the-art methods on standard databases with only a handful of labeled examples.

Sequential Dual Deep Learning with Shape and Texture Features for Sketch Recognition

Abstract: Recognizing freehand sketches with high arbitrariness is greatly challenging. Most existing methods either ignore the geometric characteristics or treat sketches as handwritten characters with fixed structural ordering. Consequently, they can hardly yield high recognition performance even though sophisticated learning techniques are employed. In this paper, we propose a sequential deep learning strategy that combines both shape and texture features. A coded shape descriptor is exploited to characterize the geometry of sketch strokes with high flexibility, while the outputs of constitutional neural networks (CNN) are taken as the abstract texture feature. We develop dual deep networks with memorable gated recurrent units (GRUs), and sequentially feed these two types of features into the dual networks, respectively. These dual networks enable the feature fusion by another gated recurrent unit (GRU), and thus accurately recognize sketches invariant to stroke ordering. The experiments on the TU-Berlin data set show that our method outperforms the average of human and state-of-the-art algorithms even when significant shape and appearance variations occur.