標題: 基於形狀描述子及部分比對之食物造型擺盤系統
Happy to Eat! Plate Presentation with Foods by Using Shape Descriptor and Partial Matching
作者: 郭曜禎
王昱舜
Kuo, Yao-Chen
Wang, Yu-Shuen
多媒體工程研究所
關鍵字: 食物擺盤;拼貼藝術;Food plate presentation;Collage
公開日期: 2017
摘要: 根據研究指出,食物擺盤中多樣的形狀以及繽紛的顏色,對於兒童的影響相較於成人來的更加顯著。食物擺盤通常是由多種食材彼此間的堆疊所構成,利用多種食材的堆疊來描述其目標圖案的輪廓及顏色,其中變化之廣使其成為研究議題。然而,食材選擇的困難以及實際擺盤中無法避免的反覆嘗試皆讓造型擺盤令人望之卻步。因此我們開發了一個系統來簡化設計食物擺盤的步驟,使用者只要選擇目標圖片以及食材,便可藉由形狀描述子以及部分比對來得到多種擺盤結果,若需要改變樣式只要更換目標圖片或所選食材即可產生不同的結果,因而降低擺盤系統的難度。
Researches have shown that good plate presentation can not only attract attention but also greatly influences the perception of taste, smell, and flavor, especially to children. Plate presentation is the combination of different kinds of foods, the target images are described by the arrangement of foods to fulfill its similarity of contour and color. The wide variety of possible combination has led to lots of research. However, the choice of food could be a tough question for users. Besides, most of the food designs involve lots of task and arranging steps which can also be a trouble for users who want to give it a try. To solve the problem, we develop a new system to simplify the process of plate design. User only needs to choose the target image and several foods, our system will compute a plate presentation according to the target image with food which were chosen automatically. User only needs to change target image and modify chosen food, it will generate different plate design according to your choice.
URI: http://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070456643
http://hdl.handle.net/11536/142667
Appears in Collections:Thesis