The paper summarizes the current research in the computervisionresearchlaboratory at the University of Michigan. The laboratory concentrates on developing generic vision algorithms for industrial applications. Gene...
详细信息
We present a novel approach to solving the trajectory planning problem (TPP) in time-varying environments. The essence of our approach lies in a heuristic but natural decomposition of TPP into two subproblems: (1) pla...
详细信息
We present a novel approach to solving the trajectory planning problem (TPP) in time-varying environments. The essence of our approach lies in a heuristic but natural decomposition of TPP into two subproblems: (1) planning a path to avoid collision with static obstacles and (2) planning the velocity along the path to avoid collision with moving obstacles. We call the first subproblem the path planning problem (PPP) and the second the velocity planning problem (VPP). Thus, our decomposition is summarized by the equation TPP right arrow PPP + VPP. The symbol right arrow indicates that the decomposition holds under certain assumptions, e.g., when obstacles are moving independently of (i.e., not tracking) the robot. Furthermore, we pose the VPP in path-time space, where time is explicitly represented as an extra dimension, and reduce it to a graph search in this space. In fact, VPP is transformed to a two-dimensional PPP in path-time space with some additional constraints. Algorithms are then presented to solve the VPP with different optimality criteria: minimum length in path-time space, and minimum time. [ABSTRACT FROM AUTHOR]
A common task in computervision is to recognize the objects in an image. Most computervision systems do this by matching models for each possible object type in turn, recognizing objects by the best matches. This is...
详细信息
A common task in computervision is to recognize the objects in an image. Most computervision systems do this by matching models for each possible object type in turn, recognizing objects by the best matches. This is not ideal, as it does not take advantage of the similarities and differences between the possible object types. The computation time also increases linearly with the number of possible objects, which can become a problem if the number is large. This paper describes a new recognition method, the feature indexed hypotheses method, which takes advantage of the similarities and differences between object types, and is able to handle cases, where there are a large number of possible object types, in sub-linear computation time. A two-dimensional occluded parts recognition system using this method is described.
This paper describes a system that applies vision techniques to extract acoustic patterns in the speech spectrogram. By processing a spectrographic image through a set of edge detectors and combining their outputs, th...
详细信息
This paper describes a system that applies vision techniques to extract acoustic patterns in the speech spectrogram. By processing a spectrographic image through a set of edge detectors and combining their outputs, the system obtains two-dimensional objects that characterize the formant patterns and general spectral properties for vowels and consonants. As a validation of the approach, a limited vowel recognition experiment was performed on the "object" spectrograms. Preliminary results show that this processing technique retains relevant acoustic information necessary to identify the underlying phonetic representation.
Orientation selection is the inference of orientation information out of images. It is one of the foundations on which other visual structures are built, since it must precede the formation of contours out of pointill...
详细信息
Orientation selection is the inference of orientation information out of images. It is one of the foundations on which other visual structures are built, since it must precede the formation of contours out of pointillist data and surfaces out of surface markings. We take a differential geometric view in defining orientation selection, and develop algorithms for actually doing it. The goal of these algorithms is formulated in mathematical terms as the inference of a vector field of tangents (to the contours), and the algorithms are studied in both abstract and computational forms. They are formulated as matching problems, and algorithms for solving them are reduced to biologically plausible terms. We show that two different matching problems are necessary, the first for 1-dimensional contours (which we refer to as Type I processes) and second for 2-dimensional flows (or Type II processes). We conjecture that this difference is reflected in the response properties of “simple” and “complex” cells, respectively, and predict several other psychophysical phenomena.
An automated system for detecting Osteogenesis Imperfecta (OI), an inheritable disorder of human connective tissue, is described. The approach is one of texture analysis, founded on standard statistical recognition of...
An automated system for detecting Osteogenesis Imperfecta (OI), an inheritable disorder of human connective tissue, is described. The approach is one of texture analysis, founded on standard statistical recognition of co-occurrence-based texture descriptors. Our contribution is to show that texture descriptors derived from gray-level co-occurrence matrices can be used in conjunction with descriptors derived from generalized co-occurrence matrices of local image features to increase performance. In fact, for the OI problem, our system demonstrates a level of performance which is significantly better than that of medical specialists.
Co-occurrence matrices are a popular representation for the texture in images. They contain a count of the number of times that a given feature (e.g., a given gray level) occurs in a particular spatial relation to ano...
Co-occurrence matrices are a popular representation for the texture in images. They contain a count of the number of times that a given feature (e.g., a given gray level) occurs in a particular spatial relation to another given feature. However, because of the large number of spatial relations that are possible within an image, heuristic or interactive techniques have usually been employed to select the relation to use for each problem. In this paper we present a statistical approach to finding those spatial (or other) relations that best capture the structure of textures when the co-occurrence matrix representation is used. These matrices should thus be well suited for discriminations that are structurally based.
There are two different sources of information in relaxation labeling processes: the initial certainty factors of the labels and the compatibility coefficients. In this paper we examine two ways in which the compatibi...
There are two different sources of information in relaxation labeling processes: the initial certainty factors of the labels and the compatibility coefficients. In this paper we examine two ways in which the compatibility coefficients influence the fixed point achieved: (1) we demonstrate how the coefficients can bias the process toward the instantiation of a subset of the labels, and (2) we show how the coefficients precisely define the set of possible fixed points. We also indicate how eigenanalysis of the derivative of a relaxation labeling process at a fixed point can be used to study the stability of the fixed point. Finally, we present an empirical comparison of two statistical interpretations of the compatibility coefficients.
Within two years, both the required algorithmic competence and the necessary integrated circuit technology will have been developed to a point where practical personal reading machines for the blind will be possible. ...
详细信息
Within two years, both the required algorithmic competence and the necessary integrated circuit technology will have been developed to a point where practical personal reading machines for the blind will be possible. In this paper, the linguistic and phonetic principles needed to convert optically recognized text to speech are discussed, and it is shown how they mirror the human cognitive ability to read aloud. A perspective on the current status and rate of progress of large scale integration technology is then used to show that economical implementations of even complex text-to-speech algorithms can be realized in the short-term future. Finally, a view of important human factors problems requiring attention is given.
The four-volume set LNCS 7724--7727 constitutes the thoroughly refereed post-conference proceedings of the 11th Asian Conference on computervision, ACCV 2012, held in Daejeon, Korea, in November 2012. The total of 22...
详细信息
ISBN:
(数字)9783642373312
ISBN:
(纸本)9783642373305
The four-volume set LNCS 7724--7727 constitutes the thoroughly refereed post-conference proceedings of the 11th Asian Conference on computervision, ACCV 2012, held in Daejeon, Korea, in November 2012.
The total of 226 contributions presented in these volumes was carefully reviewed and selected from 869 submissions. The papers are organized in topical sections on object detection, learning and matching; object recognition; feature, representation, and recognition; segmentation, grouping, and classification; image representation; image and video retrieval and medical image analysis; face and gesture analysis and recognition; optical flow and tracking; motion, tracking, and computational photography; video analysis and action recognition; shape reconstruction and optimization; shape from X and photometry; applications of computervision; low-level vision and applications of computervision.
暂无评论