Codes for visual shape perception and the contribution of depth information

视觉形状感知的代码和深度信息的贡献

• 批准号: 203613-2008
• 负责人: Arguin, Martin
• 金额: $ 1.6万
• 依托单位: Université de Montréal
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2009
• 资助国家: 加拿大
• 起止时间: 2009-01-01 至 2010-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=424109
• 关键词: Codes; visual; shape; perception; contribution

Visual shape perception is fundamental for effective interactions with our environment and to recognize objects. This proposal is part of our ongoing long-term project of uncovering how the human visual system processes and represents shape information. Our previous studies have shown that two representation formats (or codes) are involved in the perception of simple shapes. One is based on global shape features such as aspect ratio, curvature, or symmetry. Another is based on the visible surfaces of the object that are delimited by its edges. Two main goals of this proposal concern these codes. Specifically, we wish to establish, for a variety of global shape properties, whether their representation is invariant to depth rotations and if so, what is the depth information required to support this invariance. Regarding the surface code, the proposed experiments will examine to what degree it is distinct from a representation of shape that exclusively rests on its edges, and whether it is invariant to rotations on the picture plane and in depth. A third main goal of this proposal is to explore the perception and representation of complex objects; i.e. those that can be segmented in two or more parts. Our previous investigation of complex objects has shown that their perception involves two separate mechanisms, one concerned with the shape of parts and another with their spatial arrangement, or configuration. We now propose to examine whether the representation of parts in complex objects uses the same global feature and surface codes we have demonstrated for simple objects. We will also examine the perception of the configuration of complex objects. Specifically, we will examine the properties of objects that make the perception of configuration more or less difficult as well as the factors that affect our capacity to discriminate objects on their configuration.

视觉形状感知对于与我们的环境有效相互作用并识别对象是基本的。该建议是我们正在进行的长期项目的一部分，即揭示人类视觉系统如何处理并代表形状信息。我们以前的研究表明，两种表示格式（或代码）参与了简单形状的感知。一种基于全局形状特征，例如纵横比，曲率或对称性。另一个基于其边缘界定对象的可见表面。该提案的两个主要目标涉及这些法规。具体来说，我们希望建立各种全球形状属性，是否对深度旋转不变，如果是，则支持支持此不变性所需的深度信息。关于表面代码，提出的实验将检查它与仅属于其边缘的形状表示的不同程度，以及它是否与图片平面上的旋转和深度之间的旋转不变。该提案的第三个主要目标是探索复杂物体的感知和表示。即那些可以分为两个或多个部分的分段的。我们先前对复杂物体的研究表明，它们的感知涉及两种独立的机制，一种涉及零件的形状，另一种涉及其空间排列或配置。现在，我们建议检查复杂对象中零件的表示是否使用我们为简单对象展示的相同的全局特征和表面代码。我们还将检查复杂对象配置的感知。具体而言，我们将检查对象的属性，这些对象的属性或多或少地使配置感知的感知以及影响我们在配置上区分对象的能力的因素。