@techreport{oai:ipsj.ixsq.nii.ac.jp:00241176, author = {今宿, 祐希 and 山肩, 洋子 and 相澤, 清晴 and Yuki, Imajuku and Yoko, Yamakata and Kiyoharu, Aizawa}, issue = {31}, month = {Nov}, note = {食事は我々の身体的・精神的健康と密接な関係がある非常に重要な活動である.口にした食事写真をアップロードするだけで栄養管理を行うアプリケーションは盛り上がりを見せる一方,精神衛生の観点を取り入れたものはほとんど無い.本研究では,GPT-4o をはじめとする大規模マルチモーダルモデルを用い,一連の食事写真記録から食事管理 QOL を推定し,メンタルヘルスの推定へと繋ぐための検討を行う., Diet is a crucial activity closely linked to both physical and mental health. While applications that allow users to manage their nutrition simply by uploading photos of their meals are gaining popularity, few incorporate aspects related to mental well-being. In this study, we aim to explore the use of Large Multimodal Models (LMMs), such as GPT-4o, to estimate the quality of life (QOL) related to dietary management based on a series of food image logs, with the ultimate goal of predicting mental health outcomes.}, title = {大規模マルチモーダルモデルを用いた食事写真記録からのメンタルヘルス推定に関する検討}, year = {2024} }