A bootstrapping algorithm for learning linear models of object classes
Published 22 November 2002
Thomas Vetter, Michael Jones, Tomaso Poggio
Citations121
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
An algorithm that automatically bootstraps the correspondence between the prototypes is described, which can be used for 2D images as well as for 3D models and is shown to synthesize successfully a flexible model of frontal face images and a flexible models of handwritten digits.
Abstract
Flexible models of object classes, based on linear combinations of prototypical images, are capable of matching novel images of the same class and have been shown to be a powerful tool to solve several fundamental vision tasks such as recognition, synthesis and correspondence. The key problem in creating a specific flexible model is the computation of pixelwise correspondence between the prototypes, a task done until now in a semiautomatic way. In this paper we describe an algorithm that automatically bootstraps the correspondence between the prototypes. The algorithm -which can be used for 2D images as well as for 3D models-is shown to synthesize successfully a flexible model of frontal face images and a flexible model of handwritten digits.
Keywords
Computer ScienceEngineering
IEEE Transactions on Pattern Analysis and Machine IntelligenceApplication of the Karhunen-Loeve procedure for the characterization of human faces
2,454 Citations1990Michael Kirby, L. Sirovich
The use of natural symmetries (mirror images) in a well-defined family of patterns (human faces) is discussed within the framework of the Karhunen-Loeve expansion, which results in an extension of the data and imposes even and odd symmetry on the eigenfunctions of the covariance matrix.
Alignment by maximization of mutual information
1,084 Citations2002Paul Viola, William M. Wells
IEEE Transactions on Pattern Analysis and Machine IntelligenceRecognition by linear combinations of models
817 Citations1991Shimon Ullman, Ronen Basri
An approach to visual object recognition in which a 3D object is represented by the linear combination of 2D images of the object is proposed and it is shown that for objects with sharp edges as well as with smooth bounding contours, the set of possible images of a given object is embedded in a linear space spanned by a small number of views.
Active Shape Models — ‘Smart Snakes’
656 Citations1992T.F. Cootes, Chris Taylor
‘Active Shape Models’ which iteratively adapt to refine estimates of the pose, scale and shape of models of image objects are described, shown to be a powerful method for refining estimates of object shape and location.
ScienceImage Representations for Visual Learning
270 Citations1996David Beymer, Tomaso Poggio
Computer vision researchers are developing new approaches to object recognition and detection that are based almost directly on images and avoid the use of intermediate three-dimensional models.
Recognition and Structure from One 2D Model View: Observations on Prototypes, Object Classes and Symmetries
183 Citations1992Tomaso Poggio, Thomas Vetter
It is proved that for any bilaterally symmetric 3D object one non- accidental 2D model view is sufficient for recognition and linear transformations can be learned exactly from a small set of examples in the case of "linear object classes".
A unified approach to coding and interpreting face images
180 Citations2002Andreas Lanitis, Chris Taylor +1 more
A compact parametrised model of facial appearance which takes into account all sources of variability and can be used for tasks such as image coding, person identification, pose recovery, gender recognition and expression recognition is described.
DSpace@MIT (Massachusetts Institute of Technology)Example Based Image Analysis and Synthesis
172 Citations1993David Beymer, Amnon Shashua +1 more
The mapping from novel images to a vector of ``pose'''' and ``expression'''' parameters can be learned from a small set of example images using a function approximation technique that it analysis network.
3D position, attitude and shape input using video tracking of hands and lips
89 Citations1994Andrew Blake, Michael Isard
The system described here can track accurately the curved silhouettes of moving non-polyhedral objects at frame-rate, for example hands, lips, legs, vehicles, fruit, and without any special hardware beyond a desktop workstation and a video-camera and framestore.
Multi-resolution search with active shape models
63 Citations2002T.F. Cootes, Chris Taylor +1 more
This work describes a multiresolution approach to image search using flexible shape models and gives results of quantitative experiments which show a significant increase in both speed, robustness and quality of fit compared to previous methods.
A Generic System for Image Interpretation Using Flexible Templates
48 Citations1992Andrew Hill, T.F. Cootes +1 more
A generic approach to image interpretation is described, based on combining a general method of building flexible template models with Genetic Algorithm (GA) search, and demonstrating that the new method offers significant improvements when compared to previously reported approaches to flexible template matching.
DSpace@MIT (Massachusetts Institute of Technology)Model-Based Matching by Linear Combinations of Prototypes
42 Citations1996Michael Jones, Tomaso Poggio
The technique can be used for a number of applications including the computation of correspondence between novel images of a certain known class, object recognition, image synthesis and image compression.
Lecture notes in computer scienceImage synthesis from a single example image
39 Citations1996Thomas Vetter, Tomaso Poggio
This work exploits 2D image transformations that are specific to the relevant object class and learnable from example views of other “prototypical” objects of the same class.
Model-based matching of line drawings by linear combinations of prototypes
35 Citations2002Michael Jones, Tomaso Poggio
A technique for finding pixelwise correspondences between two images by using models of objects of the same class to guide the search, which applies to line drawings of objects.
A system of analyzing and synthesizing facial images
28 Citations1991Chang Soo Choi, T. Okazaki +2 more
A method which analyzes and synthesizes the facial images on the basis of a three-dimensional facial shape model is presented, extended so that the features of parts as well as the whole of the face can be analyzed and synthesized.
DSpace@MIT (Massachusetts Institute of Technology)The photo-realistic synthesis of novel views from example images
1 Citations1996Stephen Lines