VISUAL ROUTINES
VISUAL ROUTINES
复制标题
DOI：
10.1016/0010-0277(84)90023-4
复制
发表时间：
1984-01-01
期刊：
影响因子：
3.4
通讯作者：
ULLMAN, S
中科院分区：
文献类型：
--
作者：
ULLMAN, S
研究了早期表征创建之外的视觉信息处理。此级别的基本要求是建立视觉抽象形状属性和空间关系的能力。这种能力在物体识别、视觉引导操作和更抽象的视觉思维中发挥着重要作用。对于人类视觉系统来说，从计算的角度来看，对空间属性和关系的感知是复杂的，但通常看起来是即时且毫不费力的，具有欺骗性。人类系统分析空间信息的熟练程度远远超过了当前人工系统的能力。因此，对这种能力基础的计算的研究可能会导致开发新的更有效的视觉信息空间分析方法。对抽象形状属性和空间关系的感知提出了根本性的困难，对视觉信息的整体处理产生了重大影响。空间关系的计算显然将视觉信息的分析分为两个主要阶段。第一个是自下而上创建可见环境的某些表示。第二阶段涉及将称为视觉例程的过程应用于第一阶段构建的表示。这些例程可以建立无法在初始表示中明确表示的属性和关系。视觉例程由基本操作序列组成。不同属性和关系的例程共享基本操作。使用一组固定的基本操作，视觉系统可以组合不同的例程来提取无限多种形状属性和空间关系。在更详细的层面上，主要基于其潜在用途，并部分得到经验证据的支持，提出了一些看似合理的基本操作。讨论的操作包括处理焦点的转移、索引到奇怪的位置、有界激活、边界跟踪和标记。简要讨论了将这些基本操作组装成有意义的视觉例程的问题。
The processing of visual information beyond the creation of the early representations is examined. A fundamental requirement at this level is the capacity to establish visually abstract shape properties and spatial relations. This capacity plays a major role in object recognition, visually guided manipulation and more abstract visual thinking. For the human visual system, the perception of spatial properties and relations that are complex from a computational standpoint nevertheless often appears deceivingly immediate and effortless. The proficiency of the human system in analyzing spatial information far surpasses the capacities of current artificial systems. The study of the computations that underlie this competence may, therefore, lead to the development of new more efficient methods for the spatial analysis of visual information. The perception of abstract shape properties and spatial relations raises fundamental difficulties with major implications for the overall processing of visual information. The computation of spatial relations apparently divides the analysis of visual information into 2 main stages. The 1st is the bottom-up creation of certain representations of the visible environment. The 2nd stage involves the application of processes called visual routines to the representations constructed in the 1st stage. These routines can establish properties and relations that cannot be represented explicitly in the initial representations. Visual routines are composed of sequences of elemental operations. Routines for different properties and relations share elemental operations. Using a fixed set of basic operations, the visual system can assemble different routines to extract in unbounded variety of shape properties and spatial relations. At a more detailed level, a number of plausible basic operations are suggested, based primarily on their potential usefulness, and supported in part by empirical evidence. The operations discussed include shifting of the processing focus, indexing to an odd-man-out location, bounded activation, boundary tracing and marking. The problem of assembling such elemental operations into meaningful visual routines is discussed briefly.