Learning attention-controllable border-ownership for objectness inference and binding

Human visual systems can parse a scene composed of novel objects and infer their surfaces and occlusion relationships without relying on object-specific shapes or textures. Perceptual grouping can bind together spatially disjoint entities to unite them as one object even when the object is entirely novel, and bind other perceptual properties like color and texture to that object using object-based attention. Border-ownership assignment, the assignment of perceived occlusion boundaries to specific perceived surfaces, is an intermediate representation in the mammalian visual system that facilitates this perceptual grouping. Since objects in a scene can be entirely novel, inferring border ownership requires integrating global figural information, while dynamically postulating what the figure is, a chicken-and egg process that is complicated further by missing or conflicting local evidence regarding the presence of boundaries. Based on neuroscience observations, we introduce a model -- the cloned Markov random field (CMRF)-- that can learn attention-controllable representations for border-ownership. Higher-order contour representations that distinguish border-ownerships emerge as part of learning in this model. When tested with a cluttered scene of novel 2D objects with noisy contour-only evidence, the CMRF model is able to perceptually group them, despite clutter and missing edges. Moreover, the CMRF is able to use occlusion cues to bind disconnected surface elements of novel objects into coherent objects, and able to use top-down attention to assign border ownership to overlapping objects. Our work is a step towards dynamic binding of surface elements into objects, a capability that is crucial for intelligent agents to interact with the world and to form entity-based abstractions.

Download Full-text

Data Storage Layout for Object-Based De-Duplication System

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.303-306.2284 ◽

2013 ◽

Vol 303-306 ◽

pp. 2284-2288

Author(s):

Fang Yan ◽

Yu An Tan

Keyword(s):

Data Storage ◽

Storage System ◽

Unstructured Data ◽

Effective Solution ◽

Advanced Method ◽

Data Object ◽

Object Based ◽

The World ◽

Long Time ◽

Reduce Energy Consumption

The world is increasingly awash in more and more unstructured data. Object-based data de-duplication is the current most advanced method and is the effective solution for detecting duplicate data. We developed an energy saving policy for conventional disk based RAID systems. According to the characteristics of object-based data de-duplication, we introduce object layout strategies for unstructured data applications; disk accesses are concentrated in a part of the disks in a long time which is conducive to scheduling other disks into standby or shutdown mode. Our proposed methods reduce energy consumption of de-duplication storage system.

Download Full-text

An objective measure of the relative strength of perceptual grouping cues using object-based attention

Journal of Vision ◽

10.1167/8.6.1019 ◽

2010 ◽

Vol 8 (6) ◽

pp. 1019-1019

Author(s):

A. S. Greenberg ◽

S. Yantis

Keyword(s):

Perceptual Grouping ◽

Objective Measure ◽

Relative Strength ◽

Object Based

Download Full-text

Semantic Segmentation of Remote Sensing Imagery Using an Object-Based Markov Random Field Model With Auxiliary Label Fields

IEEE Transactions on Geoscience and Remote Sensing ◽

10.1109/tgrs.2017.2658731 ◽

2017 ◽

Vol 55 (5) ◽

pp. 3015-3028 ◽

Cited By ~ 14

Author(s):

Chen Zheng ◽

Yun Zhang ◽

Leiguang Wang

Keyword(s):

Remote Sensing ◽

Random Field ◽

Markov Random Field ◽

Field Model ◽

Semantic Segmentation ◽

Random Field Model ◽

Remote Sensing Imagery ◽

Markov Random Field Model ◽

Markov Random ◽

Object Based

Download Full-text

The Design and Architecture of a Secure Agent Transport Protocol for E-Commerce

Electronic Commerce ◽

10.4018/978-1-878289-76-6.ch020 ◽

2011 ◽

pp. 321-336 ◽

Cited By ~ 5

Author(s):

Yang Yang ◽

Sheng-Uei Guan

Keyword(s):

World Wide Web ◽

Electronic Commerce ◽

Intelligent Agents ◽

Credit Card ◽

World Wide ◽

Transport Protocol ◽

Primary Concern ◽

Security Framework ◽

Computer Agent ◽

The World

With the proliferation of Internet, electronic commerce (e-commerce) is beginning to take the center stage in the commerce world. Transactions via electronic means have been growing rapidly over recent years, both in terms of turnover amount and volume. It is estimated that the trend will continue, as more and more businesses have already started or have plans to put their products/services online. However, the development of e-commerce is hindered by several factors. One of them is the lack of intelligence. Today, there is little intelligence in the World Wide Web. Users cannot delegate jobs to ‘agents’ that autonomously perform the desired tasks for their owners. One way to resolve this is through the introduction of ‘smart software programs’, or intelligent agents. With an agent architecture in place, users can delegate tasks to agents. An agent can help its owner to search for and filter information, negotiate with other agents, and even perform transactions on behalf of its owner. It is predicted that agent usage will become the mainstream in the future, not just in the field of e-commerce, but in the World Wide Web as well (Guilfoyle, 1994; Corley, 1995). Due to the nature of e-commerce, security becomes a primary concern for any architecture under this category. In fact, the threats to e-commerce come mostly from the area of security. Credit card companies lose billions of dollars every year on card frauds. Bank networks are broken into and millions are transferred out without the administration’s immediate knowledge. In order to fight against these electronic crimes, it is necessary to protect our architecture with a solid security framework. Besides the security needs, it is desirable for agents to have roaming capability as well. Roaming extends the agent’s capability well beyond the limitations imposed by its owner’s computer. Agent operations should not be affected by factors such as the availability of network, the limitation on bandwidth, or the lack of computing resources. Roaming agents should be able to physically leave their owners’ machines and perform their operations using the computing resources on hosting machines.

Download Full-text

The Exploration of Autonomous Vehicles

10.4018/978-1-6684-3694-3.ch046 ◽

2022 ◽

pp. 930-944

Author(s):

Anthony J. Gephardt ◽

Elizabeth Baoying Wang

Keyword(s):

Neural Networks ◽

Deep Learning ◽

Intelligent Agents ◽

Autonomous Vehicles ◽

Learning Processes ◽

Artificial Intelligent ◽

The World ◽

History Of ◽

The Creation

This chapter explores the world of autonomous vehicles. Starting from the beginning, it covers the history of the automobile dating back to 1769. It explains how the first production automobile came about in 1885. The chapter dives into the history of auto safety, ranging from seatbelts to full-on autonomous features. One of the main focuses is the creation and implementation of artificial intelligent (AI), neural networks, intelligent agents, and deep Learning Processes. Combining the hardware on the vehicle with the intelligence of AI creates what we know as autonomous vehicles today.

Download Full-text

A Cooperative Communicative Intelligent Agent Model for E-Commerce

Managing E-Commerce and Mobile Computing Technologies ◽

10.4018/978-1-93177-746-9.ch015 ◽

2011 ◽

pp. 208-225

Author(s):

Ric Jentzsch ◽

Renzo Gobbin

Keyword(s):

Information Systems ◽

Intelligent Agents ◽

World Wide ◽

Intelligent Agent ◽

Business Information Systems ◽

Agent Model ◽

Business Operations ◽

Business Information ◽

The World ◽

Modern Business

The complexities of business continue to expand. First technology, then the World Wide Web, ubiquitous commerce, mobile commerce, and who knows. Business information systems need to be able to adjust to these increased complexities, while not creating more problems. Here, we put forth a conceptual model for cooperative communicative intelligent agents that can extend itself to the logical constructs needed by modern business operations today and tomorrow.

Download Full-text

A Theory of Physically Embodied and Causally Effective Agency

Information ◽

10.3390/info9100249 ◽

2018 ◽

Vol 9 (10) ◽

pp. 249 ◽

Cited By ~ 1

Author(s):

Kathryn Blackmond Laskey

Keyword(s):

Quantum Theory ◽

Intelligent Agents ◽

Alternative Hypothesis ◽

Causal Knowledge ◽

Scientific Evaluation ◽

Physical Processes ◽

Realistic Interpretation ◽

The World ◽

Mathematical Formalization ◽

Embodied Agent

Causality is fundamental to agency. Intelligent agents learn about causal relationships by interacting with their environments and use their causal knowledge to choose actions intended to bring about desired outcomes. This paper considers a causal question that is central to the very meaning of agency, that of how a physically embodied agent effects intentional action in the world. The prevailing assumption is that both biological and computer agents are automatons whose decisions are determined by the physical processes operating in their information processing apparatus. As an alternative hypothesis, this paper presents a mathematical model of causally efficacious agency. The model is based on Stapp’s theory of efficacious choice in physically embodied agents. Stapp’s theory builds on a realistic interpretation of von Neumann’s mathematical formalization of quantum theory. Because it is consistent with the well-established precepts of quantum theory, Stapp’s theory has been dismissed as metaphysical and unfalsifiable. However, if taken seriously as a model of efficacious choice in biological agents, the theory does have empirically testable implications. This paper formulates Stapp’s theory as an interventionist causal theory in which interventions are ascribed to agents and can have macroscopically distinguishable effects in the world. Empirically testable implications of the theory are discussed and a path toward scientific evaluation is proposed. Implications for artificial intelligence are considered.

Download Full-text

The Ternus Effect

10.1093/acprof:oso/9780199794607.003.0099 ◽

2017 ◽

Cited By ~ 1

Author(s):

Elisabeth Hein

Keyword(s):

Twentieth Century ◽

Eye Movements ◽

Visual System ◽

Apparent Motion ◽

Perceptual Grouping ◽

The Other ◽

The World ◽

Motion Display ◽

In The Beginning ◽

Insight Into

The Ternus effect refers to an ambiguous apparent motion display in which two or three elements presented in succession and shifted horizontally by one position can be perceived as either a group of elements moving together or as one element jumping across the other(s). This chapter introduces the phenomenon and describes observations made by Pikler and Ternus in the beginning of the twentieth century. Next, reasons for continued interest in the Ternus effect are discussed and an overview of factors that influence it offered, including low-level image-based factors, for example luminance, as well as higher-level scene-based factors, for example perceptual grouping. The chapter ends with a discussion of theories regarding the mechanisms underlying the Ternus effect, providing insight into how the visual system is able to perceive coherent objects in the world despite discontinuities in the input (e.g., as a consequence of eye movements or object occlusion).

Download Full-text