Computational aesthetics analyzes various expressions arising in art and music works from an aesthetic perspective, and strives to utilize them for the purpose of re-synthesis. Evaluation of such diverse expressions, including exaggeration and omission, is not unified; it does change depending on the intention of the creator and the taste of a person who appreciates the work. This team pursues computational approaches to identification of aesthetic elements.


:April enrollment:September enrollment:Team leader

Research themes

Spectrail (2016―)
この画像には alt 属性が指定されていません

Visual Analytics for Compositional Process with Sound Analysis and Synthesis

LinDA (2018―)
この画像には alt 属性が指定されていません

A Semi Automatic System for Generating Cityscape Background Images

Character recognition of seal scripts (2019―)
この画像には alt 属性が指定されていません

Script Translation in Calligraphic Works Using Deep Learning

Automatic Generation of 3D Anime-like NPC (2019―)
この画像には alt 属性が指定されていません

Automatic Generation of 3D Natural Anime-like Non-Player Characters with Machine Learning

Pottery Wheel-type Transparent Canvas (2020―)
この画像には alt 属性が指定されていません

An Interface for Drawing Effects on 3D Pottery Wheel-type Transparent Canvas