Abstract
Aspects of the present disclosure are directed to mobile robotic devices that are capable of reproducing captured imagery (e.g., images captured by a camera that is on-board the robotic device) through the use of device locomotion. For example, the robotic device can capture (or be operated to capture) an image (e.g., a self-portrait photograph which is also known as a “selfie”) and can then reproduce the image (e.g., a stylized version thereof) by controlling motion of at least a part of the robotic device to draw at least a portion of the captured scene onto a medium. For example, the robotic device can move a portion of the device that holds a pen, pencil, marker, paintbrush, etching tool, and/or the like to draw the scene onto a piece of paper, wood, canvas, glass, metal, etc. In addition, in some implementations, machine learning techniques such as “style transfer” techniques can be used to stylize the captured imagery into a particular style (e.g., a pencil sketch style portrait) prior to reproduction.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Daniels, Melissa and Bae, Ryan, "Reproduction of Captured Imagery via Machine Learning and Device Locomotion", Technical Disclosure Commons, (March 04, 2019)
https://www.tdcommons.org/dpubs_series/2002