Automated Facial Expression Analysis for Research and Clinical Use

用于研究和临床用途的自动面部表情分析

• 批准号: 8270831
• 负责人: JEFFREY F COHN
• 金额: $ 62.14万
• 依托单位: UNIVERSITY OF PITTSBURGH AT PITTSBURGH
• 依托单位国家: 美国
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-05-01 至 2017-02-28
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8270831
• 关键词: Address; Algorithms; Anxiety; Behavior; Behavioral Sciences; Benchmarking; Bipolar Disorder; Clinical; Code; Communities; Computational algorithm; Computer Vision Systems; Computer software; Computer-Assisted Image Analysis; Data; Data Set; Databases; Detection; Development; Disease; Emotions; Employee Strikes; Environment; Etiology; Face; Facial Expression; Gender; Generic Drugs; Head; Hour; Human; Imagery; Instruction; Interview; Interviewer; Label; Laboratories; Learning; Lighting; Machine Learning; Major Depressive Disorder; Manuals; Measurement; Measures; Methods; Modeling; Motion; Movement; Nature; Pain; Participant; Personal Computers; Persons; Process; Psychopathology; Reporting; Research; Research Personnel; Running; Schizophrenia; Signal Transduction; Structure; System; Techniques; Testing; Time; Training; Treatment outcome; United States National Institutes of Health; Work; addiction; base; clinical practice; follow-up; insight; meetings; nervous system disorder; positive emotional state; skin color; suicidal risk; tool

DESCRIPTION (provided by applicant): Facial expression has been a focus of emotion research for over a hundred years. In recent decades observations of facial expressions have yielded critical and dramatic insights about the etiology of psychopathology, and have proven capable of predicting treatment outcomes (see Ekman & Rosenberg, 2005). Despite these original striking findings, there has been surprisingly little follow-up work. The primary reason fr the lack of sustained research is that the most reliable manual systems for measuring facial expression often require considerable training and are labor intensive. Automated measurement using computer vision and machine learning seeks to address the need for valid, efficient, and reproducible measurement. Recent systems have shown promise in fairly small studies using posed behavior or structured contexts with confederates, or trained interviewers, or pre-trained (person-specific) face models. For automated coding to be applied in real-world settings, a large data base with ample variability in pose, head motion, skin color, gender, partial occlusion, and expression intensity is needed. We have developed a unique database that meets this need and the algorithms necessary to enable robust automated coding. The database consists of 720 participants in three-person groups engaged in a group formation task. In a preliminary study, we demonstrated that our algorithms can successfully code two key facial signals associated with human emotion in this relatively unconstrained context (Cohn & Sayette, 2010). To achieve efficient, accurate, and valid measurement of facial expression usable in research and clinical settings, we aim to 1) train and validate classifiers to achieve reliable facial expression detectin across this unprecedentedly large, diverse data set; 2) extend the previous person-specific methods to person-independent (generic) facial feature detection, tracking, and alignment; and 3) make these tools available for research and clinical use. PUBLIC HEALTH RELEVANCE: The project has two target application domains. For behavioral science, automated facial expression analysis will provide researchers with powerful tools to examine basic questions in emotion and interpersonal processes, as well as emotion processes underlying diverse forms of psychopathology and neurologic disorder. For clinical use, automated facial expression analysis will help clinicians to assess vulnerability and protective factors and objectively evaluate course of treatment across a wide range of disorders including major depression, bipolar disorder, schizophrenia, anxiety, addiction, suicide risk, and pain.

描述（由申请人提供）：面部表情一直是情绪研究的焦点超过一百年。近几十年来，对面部表情的观察已经产生了关于精神病理学病因学的关键和戏剧性的见解，并且已经证明能够预测治疗结果（参见Ekman & Rosenberg，2005）。尽管有这些最初的惊人发现，但令人惊讶的是，后续工作很少。缺乏持续研究的主要原因是，最可靠的测量面部表情的手动系统通常需要大量的培训，而且劳动密集型。使用计算机视觉和机器学习的自动化测量旨在解决有效，高效和可重复测量的需求。最近的系统已经在相当小的研究中显示出了希望，这些研究使用了构成的行为或结构化的上下文与同盟者，或经过训练的面试官，或预先训练的（特定于个人的）面部模型。为了在真实世界的设置中应用自动编码，需要在姿势、头部运动、肤色、性别、部分遮挡和表达强度方面具有充分可变性的大型数据库。我们已经开发了一个独特的数据库，以满足这一需求和必要的算法，使强大的自动编码。该数据库由720名参与者组成，他们分成三人小组，参加一个小组的形成任务。在一项初步研究中，我们证明了我们的算法可以在这种相对不受约束的环境中成功地编码与人类情感相关的两个关键面部信号（Cohn & Sayette，2010）。为了实现研究和临床环境中可用的高效、准确和有效的面部表情测量，我们的目标是：1）训练和验证分类器，以在这个前所未有的大而多样的数据集上实现可靠的面部表情检测; 2）将以前的个人特定方法扩展到与个人无关（通用）的面部特征检测、跟踪和对齐;以及3）使这些工具可用于研究和临床使用。 公共卫生相关性：该项目有两个目标应用领域。对于行为科学，自动面部表情分析将为研究人员提供强大的工具，以检查情绪和人际过程中的基本问题，以及各种形式的精神病理学和神经系统疾病的情绪过程。对于临床使用，自动面部表情分析将帮助临床医生评估脆弱性和保护因素，并客观地评估各种疾病的治疗过程，包括重度抑郁症，双相情感障碍，精神分裂症，焦虑，成瘾，自杀风险和疼痛。