3D Sketch Recognition Using The Microsoft Kinect

The concept of sketch-based recognition has recently been used to enhance object categorization and speed up image retrieval. However, in each of the previous studies, the user was required to sketch on a two-dimensional plane. Currently there haven’t been any studies on the performance of incorporating depth information into a sketch. The motivation behind this project is to develop software that will allow a user to draw in a three-dimensional space, incorporating this information, as well as determining whether this depth information will result in higher accuracies for object categorization. First, utilizing the Microsoft Kinect, software was developed to establish a virtual drawing board for three-dimensional sketching. Second, a new learning-based approach is proposed to allow for model feature extraction and recognition. The experimental results demonstrate the validity of the study as well as the effectiveness of the proposed solution.

Name thesis:degree_name

Master of Science

Level thesis:degree_level

Masters

Discipline thesis:degree_discipline

Computer Science

Grantor

Texas State University

Year dc:date.issued

2014

Author dc:creator

• Bulgerin, Travis

Advisor dc:contributor.advisor

• Lu, Yijuan

Committee members dc:contributor.committeemember

• Ngu, Anne H.H.
• Zong, Ziliang

• sketchs
• recognition
• Microsoft

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10877/4988

OAI identifier oai:identifier

oai:digital.library.txst.edu:10877/4988