Drawing Robot

The project REAP (Robotic Expression of Acquired Penmanship) is dedicated to exploring the use of robots and machine intelligence algorithms as a novel medium for artistic expression. In this regard, most art-producing devices and systems reported to-date are designed to perform free-form rendition of western style artworks. When the character to be recognized as image pattern and the information are given such as the brush stroke trajectory, width, junction area, the system starts to find a good approach to replicate the character. These features, together with writing order information, are sent to the robot to write character as the human calligrapher does. The aim of this work is, firstly, to teach calligraphy skills to the robot, and secondly in turn, in order to preserve the character calligraphy culture, have the robot teach the beginner calligraphy skills. For painting aspect, it is capable to draw and sketch line under the features given above. There are also several topics are related to the project.

Artistic rendering with Drawing Robot
 with a 2-D source image as input, e.g. a photograph of somebody or something, the robot first analyzes the image and figures out an digital artistic image on computer using NPR (Non-Photorealistic Rendering) techniques. Then the robot draws stroke by stroke on real canvas according to the digital image gotten previously. In order to perfect the real rendering, vision feedback is used to compare the real image and the source one, and refine the real output according to the difference. 
 Artistic style can include pen-and-ink (current work), oil painting, and water color and so on. And the input can also be a 3-D model of something, and in fact this is more helpful for our work since 3-D can provide more information to us. However, 3-D information is usually more difficult for us to get. Anyway, my research aims to help people, who feel hard to make artistic drawing by themselves, to create artistic works of their own.

Related publication
Josh H. M. Lam, Y. Yam, “A Skeletonization Technique Based on Delaunay Triangulation and Piecewise Bezier Interpolation”,
The 6th IEEE International Conference on Information, Communications and Signal Processing, Singapore, December, 2007.
Josh H. M. Lam, K. W. Lo, Y. Yam, “Robot Drawing Techniques for Contoured Surface Using an Automated Sketching Platform”,
The 3rd Annual IEEE Conference on Automation Science and Engineering, Scottsdale, Arizona, USA, September, 2007