The Object Data Bank: A Collaborative Project Proposal to Provide a Standardized Realistic Stimulus Set of Common Objects for Experimental Psychology

对象数据库：为实验心理学提供一组标准化现实刺激的常见对象的合作项目提案

• 批准号: 9412456
• 负责人: Michael Tarr
• 金额: $ 6.72万
• 依托单位: Yale University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1994
• 资助国家: 美国
• 起止时间: 1994-09-15 至 1996-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9412456&HistoricalAwards=false
• 关键词: Object; Data; Bank; Collaborative; Project

TARR This object databank is intended as a collaborative project among experimental psychologists to develop a set of standardized stimulus materials depicting familiar common objects. The ultimate goal of the project is to make available to the research community a set of 300-500 photorealistically illuminated and shaded objects. Each object will be developed using 3-dimensional computer-aided design (CAD) modeling techniques and will be rendered using state-of-the-art raytracing. A given object will be available in several standard formats for researchers to use in generating their own stimulus images without the need of duplicating the time-intensive modeling required for even a single object. To facilitate researchers who need only high quality images but not specific control of lighting, textures and colors, objects will also be available as collections of already rendered images, each object depicted in realistic colors at a minimum of 12 3-dimensional viewpoints generated by 30 degree rotations around the vertical axis. The set of objects will initially be based on the widely used set of 2-dimensional line drawings of objects introduced and distributed by Snodgrass and Vanderwart (1980). This set of familiar common objects has been used in many different experimental settings and has provided ease of stimulus preparation and some stimulus standardization between differing research paradigms. The goal is to begin by creating 3-dimensional analogs of each item in the Snodgrass and Vanderwart set -- this provides the added advantage of offering some previously collected norm data about object categories, as well as some basis for comparison with earlier studies. To illustrate some of the many potential applications of this new set of objects, several proposed experiments are included from the areas of object perception , visual memory, scene recognition, neuropsychology of object recognition, neuropsychology of memory, clinical assessment of attentional disorders. Many of the proposed st udies would be impossible without realistic images of common objects. It is expected that the stimulus set will have far wider applications than even this broad range of disciplines, the common theme being that highly realistic visual stimuli can only enhance the ecological relevance of current paradigms, as well as make available new paradigms that were previously unrealizable. ***

这个对象数据库是实验心理学家之间的一个合作项目，旨在开发一套标准化的刺激材料，描述熟悉的常见物体。该项目的最终目标是为研究界提供一组300-500个逼真的照明和阴影物体。每个对象都将使用三维计算机辅助设计（CAD）建模技术进行开发，并将使用最先进的光线追踪技术进行渲染。给定的对象将以几种标准格式提供，供研究人员用于生成自己的刺激图像，而无需复制甚至单个对象所需的耗时建模。为了方便那些只需要高质量图像而不需要对光线、纹理和颜色进行特定控制的研究人员，对象也将作为已经渲染图像的集合提供，每个对象都以逼真的颜色描绘，每个对象都是通过绕垂直轴旋转30度生成的至少12个三维视点。该对象集最初将基于Snodgrass和Vanderwart（1980）引入并分发的广泛使用的对象二维线条图集。这组熟悉的常见对象已在许多不同的实验环境中使用，并为刺激准备提供了便利，并在不同的研究范式之间提供了一些刺激标准化。我们的目标是首先创建Snodgrass和Vanderwart集合中每个项目的三维类似物——这提供了提供一些先前收集的关于对象类别的规范数据的额外优势，以及与早期研究进行比较的一些基础。为了说明这组新对象的一些潜在应用，几个拟议的实验包括从物体感知、视觉记忆、场景识别、物体识别的神经心理学、记忆的神经心理学、注意力障碍的临床评估等领域。如果没有普通物体的真实图像，许多拟议的研究是不可能的。预计刺激集的应用范围将远远超过这一广泛的学科范围，共同的主题是高度逼真的视觉刺激只能增强当前范式的生态相关性，以及提供以前无法实现的新范式。＊＊＊