A spatio-temporal probabilistic modeling of dynamic facial expressions

2018 Fiscal Year Final Research Report

• Project/Area Number: 16K04437
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Experimental psychology
• Research Institution: Osaka Electro-Communication University
• Principal Investigator: Komori Masashi 大阪電気通信大学, 情報通信工学部, 教授 (60352019)
• Co-Investigator(Kenkyū-buntansha): 池田 功毅 中京大学, 心理学部, 助教 (20709240)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: 表情表出 / 感情 / モーションキャプチャ

Outline of Final Research Achievements

Affective facial expressions are dynamic in nature. Several researchers have claimed that dynamically changing facial configurations plays an important role in the perception of facial patterns and in the attribution of emotional category labels. This study aimed at finding the spatio-temporal structures of spontaneous dynamic facial expressions based on hierarchical Bayesian modeling. We empirically examined the relationship between the sequential configurations of the spontaneous facial expressions and the emotional state of the facial owners. The results suggest that multiple dynamic components with various temporal characteristics appear simultaneously in the natural facial expressions and that the combination of them corresponded to the emotion states.

Free Research Field

実験心理学

Academic Significance and Societal Importance of the Research Achievements

従来の表情研究で対象とされてきた「表情」の研究では，顔面形状の変化が最大となる瞬間の顔面形状が，感情カテゴリと対応する典型的な表情として利用されてきた．すなわち，多くの従来研究は暗黙に静的な表情表象の存在を前提としていたといえる．一方，本研究ではより自然な表情に着目し，感情状態と動的な顔面形状変化の関係を記述する確率的表情表出モデルの構築を目指したものであった．これまでの研究の結果から，表情表出時の顔面形状には，時空間的な特徴が異なる複数の要素が重畳し，急速に変化する成分と持続的に変化する成分が同時に現れることが示された．