Computer vision theory: The lack thereof

Computer Vision Graphics and Image ProcessingPublished 1 November 1986

Robert M. Haralick

Citations54

TL;DR

A theoretical approach and problem statement to 3D object matching through a sensor projection, a Bayesian approach to the robust estimation of camera parameters, and to the hypothesis verification problem are outlined.

Abstract

Abstract The first part of the paper presents a brief description of how science and the scientific method might apply to computer vision and then contrasts this perspectives with three strong papers in the pose detection area of computer vision. The contrast illuminates the shortcomings and the lack of a completed computer vision theory in these papers. The second part of the paper outlines a theoretical approach and problem statement to 3D object matching through a sensor projection, a Bayesian approach to the robust estimation of camera parameters, and to the hypothesis verification problem.

Keywords

Computer ScienceEngineering

Communications of the ACMRandom sample consensus

25,463 Citations1981Martin A. Fischler, Robert C. Bolles

New results are derived on the minimum number of landmarks needed to obtain a solution, and algorithms are presented for computing these minimum-landmark solutions in closed form that provide the basis for an automatic system that can solve the Location Determination Problem under difficult viewing.

View PDF

DSpace@MIT (Massachusetts Institute of Technology)MACHINE PERCEPTION OF THREE-DIMENSIONAL SOLIDS,

1,463 Citations1963Lawrence G. Roberts

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering, 1963.

View PDF

Artificial IntelligenceIncreasing tree search efficiency for constraint satisfaction problems

1,114 Citations1980Robert M. Haralick, Gordon L. Elliott

Analytically and experimentally show that a lookahead procedure called forward checking which employs the most likely to fail principle performs better than standard backtracking, Ullman's, Waltz's, Mackworth's, and Haralick's discrete relaxation in all cases tested, and better than Gaschnig's backmarking in the larger problems.

Journal of the Franklin InstituteOptical and electro-optical information processing

706 Citations1966E. B. Brown

IEEE Transactions on Pattern Analysis and Machine IntelligenceMatching Images to Models for Registration and Object Detection via Clustering

306 Citations1982George Stockman, Steven Kopstein +1 more

A new technique for matching image features to maps or models which forms all possible pairs of image features and model features which match on the basis of local evidence alone and which is robust with respect to changes of image orientation and content.

IEEE Transactions on ComputersA Model-Based Vision System for Industrial Parts

285 Citations1978Perkins

A vision system has been developed which can determine the position and orientation of complex curved objects in gray-level noisy scenes and organizes and reduces the image data from a digitized picture to a compact representation having the appearance of a line drawing.