Computing with Connections in Visual Recognition of Origami Objects

@article{Sabbah1985ComputingWC,
  title={Computing with Connections in Visual Recognition of Origami Objects},
  author={Daniel Sabbah},
  journal={Cogn. Sci.},
  year={1985},
  volume={9},
  pages={25-50}
}
  • D. Sabbah
  • Published 1 May 1988
  • Computer Science
  • Cogn. Sci.
This paper summarizes our initial foray in tackling Artificial Intelligence problems using a connectionist approach. The particular task chosen was the visual recognition of objects in the Origami world as defined by Kanade (1978) . The two major questions answered were how to construct a connectionist network to represent and recognize projected line drawings of Origami objects and what advantages such an approach would have. The structure of the resulting connectionist network can be… 

Evidence integration for 3D object recognition: a connectionist framework

TLDR
The authors present recent work on a vision system designed to recognize 3-D objects in a depth map that has added the ability to find parametric surface intersection curves and use these disparate types of information to index into an object database.

Visual recognition using concurrent and layered parameter networks

TLDR
A vision system to recognize 3-D objects is presented, featuring a novel notion of generalized feature that allows fundamentally different types of features to interact in a coherent way, utilizing redundancies for more robust recognition.

A Complete and Extendable Approach to Visual Recognition

TLDR
A framework for 3D object recognition is presented and a complete, integrated, and implemented system that extracts planar surfaces, patches of quadrics of revolution, and planar intersection curves of these surfaces from a depth map viewing 3D objects is described.

Evidence Integrat ion for 3D Ob1 'ect Recnpnition; A Connectionist Framework

TLDR
The ability to find parametric surface intersection curves and use these disparate types of information to index into an object database is added to a vision system designed to recognize 3D objects in a depth map.

A Homogeneous Framework for Visual Recognition

TLDR
A homogeneous paradigm for evidence integration is presented, and a vision system to recognize 3D objects is demonstrated using this paradigm, which allows the system to completely avoid the common weak point of explicit, low-level scene segmentation.

Extraction Of Surface Parameters From Depth Maps Viewing Planes And Quadrics Of Revolution

This paper presents a first step in the definition and construction of a vision system to recognize three-dimensional objects whose surface are composed of planar patches and patches of quadrics of

Evidence Fusion Using Constraint Satisfaction Networks

TLDR
This paper describes a modular and coherent approach to 3D object recognition that uses constraint satisfaction networks, motivated by connectionist networks, and proposes solutions that will keep the size of the networks from growing beyond control.

A Model Of Visual Recognition Implemented Using Neural Networks

TLDR
Experimental results show that the implementation of the prototype model is able to achieve a 100% successful recognition rate with performance that degrades gracefully, and it is flexible enough to categorise data from any domain.

A recognition network model-based approach to dynamic image understanding

TLDR
It is demonstrated that gain improvement in computation time is a major reason for dynamic understanding, in this case the computational complexity in this system is polynomial when using the hypotheses tracking mechanism.

Feature Recognition Using Correlated Information Contained in Multiple Neighborboods

TLDR
A generalized neighborhood concept is presented that allows parameter-extraction operators to use the joint information of different portions of the same feature to produce up to a few orders of magnitude improvement in accuracy (signal/noise ratio) and a smoother response of the transform.

References

SHOWING 1-10 OF 47 REFERENCES

A connectionist approach to visual recognition

TLDR
A massively parallel, connectionist approach is brought to bear on the problem of visual recognition that eliminates the need for establishing a search order on exploring interpretations and exhibits many similarities with the structure and behavior of animal vision systems.

A Theory of Origami World

Representing Visual Information

TLDR
This essay surveys recent work in vision at M.I.T. from a perspective in which the representational problems assume a primary importance.

Shape Representation in Parallel Systems

TLDR
This paper considers the difficulties involved in representing shapes in parallel systems, and suggests ways of overcoming them, and provides a mechanism for shape perception and visual attention which allows a novel interpretation of the Gestalt slogan that the whole is more than the sum of its parts.

Design Of A Highly Parallel Visual Recognition System

TLDR
Relaxation in a conceptual hierarchy is proposed as a useful principle in the design of a visual recognition system that combines uniform approach to intermediate and low level vision, inherent parallel approach and sharing of partial results to dynamically cut down the search for a correct match.

MSYS: A System for Reasoning About Scenes.

TLDR
MSYS is a system for reasoning with uncertain information and inexact rules of inference that contributes to the interpretation of visual features in scene analysis by repeatedly applying rules until a consistent set of likelihood values is attained.

Shape from Texture

TLDR
For natural textures, it is shown that the uniform density assumption (texels are uniformly distributed) is enough to recover the orientation of a single textured plane in view, under perspective projection.

Recognition of 3-D Objects Using the Extended Gaussian Image

TLDR
An algorithm for reconstruction of the original shape of a convex polyhedron from its EGI based on the analysis-by-synthesis method is proposed and will apply a matching function to ESls of a candidate set for a final decision.

Connectionist Models and Their Properties

Much of the progress in the fields constituting cognitive science has been based upon the use of explicit information processing models, almost exclusively patterned after conventional serial