Modeling Color Appearance in Augmented Reality

Download or read book Modeling Color Appearance in Augmented Reality written by Nargess Hassani and published by . This book was released on 2019 with total page 128 pages. Available in PDF, EPUB and Kindle. Book excerpt: "Augmented reality (AR) is a developing technology that is expected to become the next interface between humans and computers. One of the most common designs of AR devices is the optical see-through head- mounted display (HMD). In this design, the virtual content presented on the displays embedded inside the device gets optically superimposed on the real world which results in the virtual content being transparent. Color appearance in see-through designs of AR is a complicated subject, because it depends on many factors including the ambient light, the color appearance of the virtual content and color appearance of the real background. Similar to display technology, it is vital to control the color appearance of content for many applications of AR. In this research, color appearance in the see-through design of augmented reality environment is studied and modeled. Using a bench-top optical mixing apparatus as an AR simulator, objective measurements of mixed colors in AR were performed to study the light behavior in AR environment. Psychophysical color matching experiments were performed to understand color perception in AR. These experiments were performed first for simple 2D stimuli with single color both as background and foreground and later for more visually complex stimuli to better represent real content that is presented in AR. Color perception in AR environment was compared to color perception on a display which showed they are different from each other. The applicability of the CAM16 color appearance model, one of the most comprehensive current color appearance models, in AR environment was evaluated. The results showed that the CAM16 is not accurate in predicting the color appearance in AR environment. In order to model color appearance in AR environment, four approaches were developed using modifications in tristimulus and color appearance spaces, and the best performance was found to be for Approach 2 which was based on predicting the tristimulus values of the mixed content from the background and foreground color."--Abstract.