Showing papers in "Image and Vision Computing in 1995"

TL;DR: A review of recent developments in the computer vision aspect of motionbased recognition and several methods for the recognition of objects and motions, including cyclic motion detection and recognition, lipreading, hand gestures interpretation, motion verb recognition and temporal textures classification are reported.

TL;DR: It is shown that good face reconstructions can be obtained using 83 model parameters, and that high recognition rates can be achieved.

TL;DR: Tests in line detection with synthetic and real-world images demonstrate the high speed and low memory usage of the new extensions of the RHT, as compared both to the basic RHT and other versions of the Hough transform.

TL;DR: A subpixel addressing mechanism (called linear interpolation) is utilized for intermediate pixel addressing in the differentiation step, which results in improved accuracy of corner localization and reduced computational complexity.

TL;DR: A new approach is described which generates vibrational modes when few example shapes are available and changes smoothly to using more statistical modes of variation when a large data set is presented.

TL;DR: The suggested approach for understanding behavioural vision to realize the relationships of perception and action builds on two earlier approaches, the Medusa philosophy and the Synthetic approach, and calls for synthesizing an artificial vision system by studying vision competences of increasing complexity and pursuing the integration of the perceptual components with action and learning modules.

TL;DR: It is argued that interpretation of a large number of gestures involves analysis of image dynamics to identify and characterize the gestural stroke, locating the stroke extrema in ordinal 3D space, and recognizing the hand pose at stroke Extrema.

TL;DR: A normalization algorithm has been developed which transforms pattern into its normal form such that it is invariant to translation, rotation, scaling and skew, and can serve as a pre-processing step in computer vision applications.

TL;DR: This paper proposes a list of such problems after a review of the current major 3D imaging modalities, and a description of the related medical needs, and presents some of the past and current work done in the research group EPIDAURE at INRIA on the following topics.

TL;DR: A variety of designs for polarization camera sensors that have been built to automatically sense partial linearly polarized light, and computationally process this sensed polarization information at pixel resolution to produce a visualization of reflected polarization from a scene, and/or a visualize of physical information in a scene directly related to sensed polarization.

TL;DR: It turns out that only aspects of the partial depth order (based on depth precedence in infinitesimal regions) are stable and features of the relief are invariants of general ‘relief preserving transformations’ that may actually scramble depth values at different locations.

TL;DR: In this article, the authors describe a new region growing method that uses a closed snake driven by a pressure force that is a function of the statistical characteristics of image data, which expands until its elements encounter pixels that lie outside user-defined limits relative to a seed region.

TL;DR: The theory of Lie groups is described, a systematic approach to the problem of extracting invariants from Lie groups that helps in tackling problems such as calibration-less vision, efficient matching, shape-from-motion, grouping, and several other problems considered crucial to intelligent vision.

TL;DR: The behaviour of structures with changing resolution in a linear diffusion scale space is examined, which includes not only periods of smooth change, but also four catastrophic changes: shoe surface, balanced saddle, double-saddle isophote and heteroclinic separatrix.

TL;DR: This work explores how global symmetry can be detected prior to segmentation and under noise and occlusion, and maximizes a global symmetry measure based on accumulated local support in the LSSF-a straight active contour model is used for this purpose.

TL;DR: A novel network initialization strategy is applied to achieve the desired complexity reduction and a method to verify and localize the hypotheses generated by the Hopfield network is presented using an efficient pose clustering algorithm.

TL;DR: The novelty of the method lies in its Bayesian modelling of relational consistency which leads to a global matching criterion with a unique mathematical structure and robustness to error, which allows it to operate effectively in a large space of relational mappings.

TL;DR: A neural network classifier based on a supervised version of Kohonen learning vector quantization (LVQ) is proposed to automatically and efficiently make classification criteria of the solder joint shapes according to their quality.

TL;DR: The resulting polynomial regression PDMs (PRPDM) perform well on the data for which the linear method failed, and can be used to model shape variability by means of point distribution models (PDM).

TL;DR: It is shown that some useful and intuitive three-dimensional information, such as relative positions of points and planes and 3D convex hulls, can be computed in the images without performing any three- dimensional reconstruction.

TL;DR: ‘steerable filters’ are used to estimate the motion of a scene using steerable filters synthesized by the interpolation of some ‘basis functions’ and ‘steered’ to arbitrary orientations.

TL;DR: A recognition system is described which thins the characters using conventional algorithms to produce 8-connected skeletons and uses a new pattern grammar to describe these characters in terms of simple string.

TL;DR: In this paper, the authors introduce an extended distance transform which may be used to capture more of the symmetries of a shape and describe the relationship of this extended distance transformation to the skeletal shape descriptors themselves, and other geometric phenomena related to the boundary of the curve.

TL;DR: The methods suggested in a previous paper for object recognition and location provide a complete solution to the problem of recognition of edge based descriptors for fixed 2D projected views of rigid objects.

TL;DR: A parallel thinning algorithm that relies on the iterated erosion of the pattern, and which proceeds from lower grey values towards higher ones until the grey-skeleton is finally obtained is presented.

TL;DR: Several aspects related to the application of active vision techniques to object recognition are discussed and the face recognition problem based on the face-space approach is considered to demonstrate the advantage of adopting an active retina in recognition tasks.

TL;DR: The curve fitting method used in this work does not require solving a system of equations, and therefore is fast, and results obtained are compared with those determined by the Laplacian of Gaussian operator.

TL;DR: A novel method for estimating motion parameters and point correspondences between 2D curves and 3D surfaces under small nonrigid deformations and demonstrates the correct correspondence recovery by tracking points on the face during each facial expression.

TL;DR: The proposed theory provides a unifying framework for a variety of such models that can be related to conventional scale-space in a one-to-one way, and may potentially solve the so-called “localisation problem” of linear scale- space.