| Summary: | 碩士 === 南臺科技大學 === 多媒體與電腦娛樂科學系 === 107 === This study collects the existing virtual fitting room examples for evaluation, and finally develops with Kinect V2 as an intermediary device. The development process collects clothing for establishing 3D models to evaluate how to build a very natural clothing model. Through constant testing, the test was carried out from the stand position, interface design, and the naturalness of the clothing. The number of test people was 200 people to find the best way to meet the demand. The commission also included three Taiwan-made brands of clothing, including: Hongyuan Industrial Co., Ltd., GENQUO, Jiayi boutique uniforms, a total of 15 fashion clothing to join the magic fitting room menu, the value-added service part in addition to printing the results of the public experience, sent E-MAIL, borrow The three fashion websites are connected by touch screen operation to complete the integration of the complete business model of the virtual fitting room. The last development is to use the 2D picture to provide the people to scan the images after painting, so that the people can experience their own creation. The clothes are tried through the entire magical fitting room.
In view of the evolution and scientific implications of the somatosensory identification technology involved in the virtual fitting room, and analyzing the technical models and benefits of the representative virtual fitting room case in the past five years, the museum’s current “magic fitting room” The advantages and disadvantages of 1.0" are analyzed to summarize the characteristics and design direction of "Magic Folding Room 2.0".
Since the Magic Dressing Room 2.0 is a gesture-operated interactive system with no physical interface, the position of the visual component is designed so that any size user can easily touch the operating element with the hand position in the system screen, so the interface The location is quite important. In addition, in order to make the system screen as simple as possible, try to reduce the time that the UI component stays on the screen, and use the "trigger mode" to make the time-dependent UI components appear instantly instead of simultaneously. The new depth camera recognition ability is greatly improved, and the height is not increased. The same people only need to change the position before and after to experience.
|
|---|
