H.K. Nishihara, "Practical real-time imaging stereo matcher," Optical Engineering,Vol. 23, No. 5, 1984, pp. 536-545.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....regions. Other traditional matching costs include normalized cross correlation [51, 93, 19] which behaves similar to sumof squared differences (SSD) and binary matching costs (i.e. match no match) 73] based on binary features such as edges [4, 50, 27] or the sign of the Laplacian [82]. Binary matching costs are not commonly used in dense stereo methods, however. Some costs are insensitive to differences in camera gain or bias, for example gradient based measures [100, 95] and non parametric measures such as rank and census transforms [129] Of course, it is also possible to ....

.... implementation is closely tied to the taxonomy presented in Section 3 and currently includes window based algorithms, diffusion algo6 Method Matching cost Aggregation Optimization SSD (traditional) squared difference square window WTA Hannah [51] cross correlation (square window) WTA Nishihara [82] binarized filters square window WTA Kass [63] filter banks none WTA Fleet et al. 40] phase none phase matching Jones and Malik [57] filter banks none WTA Kanade [58] absolute difference square window WTA Scharstein [95] gradient based Gaussian WTA Zabih and Woodfill [129] rank transform ....

H. K. Nishihara. Practical real-time imaging stereo matcher. Optical Engineering, 23(5):536--545, 1984.

....during aggregation. Other traditional matching costs include normalized cross correlation [34, 60, 15] which behaves similar to sum of squared differences (SSD) and binary matching costs (i.e. match no match) 46] based on binary features such as edges [33] or the sign of the Laplacian [51]. Binary matching costs are not commonly used in dense stereo methods, however. Some costs are insensitive to differences in camera gain or bias, for example gradient based measures [61] and nonparametric measures, such as rank and census transforms [82] Of course, it is also possible to ....

H. K. Nishihara. Practical real-time imaging stereo matcher. Optical Engineering, 23(5):536-545, 1984.

....The baseline of a stereo pair is the distance 16 between the camera locations of the two images. Disparity refers to the difference in image location between corresponding features in the two images, which is projectively related to the depth of the feature in the scene. Years of research (e.g. [3, 10, 17, 22, 26, 32, 35]) have shown that determining stereo correspondences by computer is difficult problem. In general, current methods are successful only when the images are similar in appearance, as in the case of human vision, which is usually obtained by using cameras that are closely spaced relative to the ....

H. K. Nishihara. Practical real-time imaging stereo matcher. Optical Engineering, 23(5):536-- 545, 1984.

.... is a vast literature on image matching, including various methods for detecting feature points [ Moravec, 1977, Dreschler and Nagel, 1982, Harris and Stephens, 1988, Noble, 1988, Rohr, 1992, Deriche and Giraudon, 1993 ] hypothesizing initial correspondences (e.g. correlation [ Gennery, 1980, Nishihara, 1984, Fua, 1993 ] feature tracking [ Mohr et al. 1992, Tomasi and Kanade, 1992, Zhang, 1994 ] and structural matching [ Barnard and Thompson, 1980, Shapiro and Haralick, 1981, Horaud and Skordas, 1989, Weng et al. 1992 ] establishing new correspondences once an initial estimate of the epipolar ....

H.K. Nishihara. PRISM, a practical real-time imaging stereo matcher. AI Memo 780, MIT, 1984.

.... is a vast literature on image matching, including various methods for detecting feature points [ Moravec, 1977, Dreschler and Nagel, 1982, Harris and Stephens, 1988, Noble, 1988, Rohr, 1992, Deriche and Giraudon, 1993 ] hypothesizing initial correspondences (e.g. correlation [ Gennery, 1980, Nishihara, 1984, Fua, 1993 ] feature tracking [ Mohr et al. 1992, Tomasi and Kanade, 1992, Zhang, 1994 ] and structural matching [ Barnard and Thompson, 1980, Shapiro and Haralick, 1981, Horaud and Skordas, 1989, Weng et al. 1992 ] establishing new cor2 respondences once an initial estimate of the ....

H.K. Nishihara. PRISM, a practical real-time imaging stereo matcher. AI Memo 780, MIT, 1984.

....optical flow. Notable is the work on implementing optical flow equations using analog devices [45] which incorporates the idea of weak continuity for preserving motion discontinuities. Others have built hardware to perform real time correlation for optical flow, depth map generation, and tracking [46, 47]. Computational models of optic flow and, in particular, the inclusion of the temporal persistence constraint, also have important implications for understanding biological vision. Tarr and Black [48] have demonstrated that temporal persistence produces systematic distortions in motion recovery ....

....to attack a difficult problem. With the purposive approach, one begins with a narrowly defined task and determines what information is both necessary and easily computable to achieve the task. This is a traditional engineering approach to robotics. In contrast, consider the approach of Nishihara [47] that on the surface appears very similar; as with the purposive approach he attempts to solve simple tasks using robust and efficient methods. Nishihara, however, begins with a biologically motivated representation based on the sign of Laplacian of Gaussian filtered images. He then defines a ....

H. K. Nishihara, Practical real-time imaging stereo matcher, Opt. Engrg. 23(5), 1984, 536-545.

....calibration process have been correctly established by a separate matching process. Let us just mention that there is a vast literature on image matching, including various methods 6 for detecting feature points [76, 18, 32, 81, 92, 15] hypothesizing initial correspondences (e.g. correlation [28, 79, 24], feature tracking [75, 108, 118] and structural matching [4, 98, 45, 115] establishing new correspondences once an initial estimate of the epipolar geometry has been obtained [117, 80] and nding and rejecting false matches using techniques from robust statistics [83, 97, 110, 109] Two recent ....

....may also be extracted from the images to be used in the rendering process. There is a tremendous amount of work in the computer vision literature on model reconstruction from intensity images with di erent paradigms such as structure from motion (e.g. 55, 108, 19, 105] shape from stereo (e.g. [70, 79]) shape from shading (e.g. 49, 84] and other model reconstruction methods (e.g. 13, 104] The extracted model may be a complete CAD model or just the positions of some geometric features like points and lines. Once the 3D model representing the scene has been reconstructed, a new view of ....

H.K. Nishihara. PRISM, a practical real-time imaging stereo matcher. AI Memo 780, MIT, 1984.

....when it was not possible to retrieve a dense and accurate reconstruction within a reasonable amount of time. # Area based: In these approaches, dense depth maps are provided by correlating the grey levels of image patches in the views being considered, assuming that they present some similarity [14, 17, 21, 40]. These methods are well adapted for relatively textured areas; however, they generally assume that the observed scene is locally fronto parallel, which causes problems for slanted surfaces and in particular near the occluding contours of the objects. Lastly, the matching process does not take ....

Keith Nishihara. Practical real-time imaging stereo matcher. Optical Engineering, 23(5), 536545, 1984. See also MIT AI MEMO-772, August 1984.

....when it was not possible to retrieve a dense and accurate reconstruction within a reasonable amount of time. Area based: In these approaches, dense depth maps are provided by correlating the grey levels of image patches in the views being considered, assuming that they present some similarity [13, 15, 16, 21, 36]. These methods are well adapted for relatively textured areas; however, they generally assume that the observed scene is locally fronto parallel, which causes problems for slanted surfaces and in particular near the occluding contours of the objects. Lastly, the matching process does not take ....

K. Nishihara, Practical real-time imaging stereo matcher, Optical Engineering 23, 1984, 536{

....variables. Keywords stereo vision, matching, correlation, frequency, scale, wavelet transform. I. Introduction T HE matching of stereoscopic images is one of the classical problems of computer vision. Several approaches have been developed (matching of characteristics or primitives [9] [11], global matching [12] object oriented matching, Correlationbased approaches are among the most popular [5] 11] but generally suffer from the lack of locality of correlation functions. The goal of this note is to present a localized version of correlation functions, based upon ....

....T HE matching of stereoscopic images is one of the classical problems of computer vision. Several approaches have been developed (matching of characteristics or primitives [9] 11] global matching [12] object oriented matching, Correlationbased approaches are among the most popular [5] [11], but generally suffer from the lack of locality of correlation functions. The goal of this note is to present a localized version of correlation functions, based upon multiresolution decompositions of the images, i.e. two dimensional wavelet transform. The localized correlation function (LCF) ....

H.K. Nishihara (1987): Practical real-time imaging stereo matcher, in Readings in Computer Vision, Fichler, M.A. and Firschein, O. Eds, Kauffman, Los Altos.

....phase of the stereo algorithm proposed by Marr and Poggio [03,04] one calculates peak values for the correlation between the borders extracted by the operator 2 G, finding the disparities from them. A subsequent interpolation phase is then necessary to get a dense depth map. Nishihara, in [07,08,09], extends Marr and Poggio work [04] He proposes to use correlation between regions in binary images instead of edges. The images are pre filtered by the 2 G and clipped to binary values (0,1) by effectively taking the sign bit of the filtered values. After correlation values between the signs ....

....require a precise (nor complete) depth reconstruction. Furthermore, a spatial adaptive technique can be aplied in a multilevel computation in order to get the desired precision near the image centers (fovea) In despite of some successfull experiences on active vision, like those presented in [07,09,10,11,12,17,18,19], the problem of matching the pixels of a stereo pair of images in real time has not been completely solved, being a critical point of the vision process. We also remark that no epipolar constraints have been used in this work. Having epipolar lines one can constrain the search for the point ....

H. K. NISHIHARA. Practical Real-Time Imaging Stereo Matcher. TR Optical Engeneering. MIT, Artificial Intelligence Laboratory. 1984

.... well known correspondence problem [17] with its equally well known difficulties (even for humans [1] 6 Related Work Much work has been done on active vision systems [3,4] and on active stereo vision in particular [11,14,21,26] Proximity spaces are themselves based on the work of Nishihara [20], which is based on the work of Marr and Poggio [19] However, such systems are seldom integrated into an agent architecture. Proximity spaces have been used to grasp weightless objects [14] and interpret 15 gestures [16] but this work marks the first time they were incorporated into a system of ....

H.K. Nishihara. Practical real-time imaging stereo matcher. Optical Engineering, 23(5):536--545, 1984.

....world points integrate light reflected off two different surface patches due to foreshortening, depth discontinuities, lens blur, and image sampling. Although some researchers have proposed measures of pixel dissimilarity that are insensitive to gain, bias, noise, and depth discontinuities [6] [9], 10] 11] 13] there seems to be no work on explicitly achieving insensitivity to image sampling. Yet this latter phenomenon can significantly change the intensity value of a pixel where the intensity function is changing rapidly and where the disparity is not an integral number of pixels ....

# H.K. Nishihara, "Practical Real-Time Imaging Stereo Matcher," Optical Eng., vol. 23, pp. 536--545, 1984.

....of CHAPTER 3. MATCHING POINTS IN IMAGES 56 edge locations for correlation. Novak [67] lists several different similarity measures and discusses their relative merits. Wong [68] describes different processing techniques that can be applied to the edges to yield more reliable matches. Nishihara [69] developed a binary correlation method based on the sign representation of the image convolved with a Laplacian of Gaussian operator, r 2 G, where G, given by G = 1 2oe e Gamma x 2 y 2 2oe 2 (3:15) is the Gaussian smoothing filter of bandwidth proportional to 1=oe. The zero crossings ....

H. K. Nishihara. A practical real-time imaging stereo matcher. Optical Engineering, 23:536--545, 1984.

....of automatically tracking pigs using simple tracking and motion capture techniques with an inexpensive off the shelf camera and framegrabber setup. Our objectives are that the resulting techniques be noise tolerant, be able to perform at a reasonable speed, be reasonably accurate and simple (Nishihara 1984). The first three criteria correspond to robustness, real time performance and sufficient accuracy with respect to a particular task. The fourth criteria, simplicity, helps in analyzing the algorithm. We explored simple tracking methods and optical flow techniques to examine if the behaviour of ....

Nishihara, H. 1984. Practical real-time imaging stereo matcher. Optical Engineering 23(5):536--553.

No context found.

H.K. Nishihara, "Practical real-time imaging stereo matcher," Optical Engineering,Vol. 23, No. 5, 1984, pp. 536-545.

No context found.

H.K. Nishihara, "Practical real-time imaging stereo matcher," Optical Engineering,Vol. 23, No. 5, 1984, pp. 536-545.

No context found.

H.K. Nishihara, "Practical real-time imaging stereo matcher," Optical Engineering,Vol. 23, No. 5, 1984, pp. 536-545.

No context found.

H.K. Nishihara. Practical real-time imaging stereo matcher. Opt. Eng., 23:536--545, Sept. 1984.

No context found.

H.K. Nishihara, "Practical real-time imaging stereo matcher," Optical Engineering,Vol. 23, No. 5, 1984, pp. 536-545.

No context found.

Nishihara HK (1984) Practical Real-Time Imaging Stereo Matcher. Opt Eng 23: 536--545

No context found.

H.K.Nishihara.Practicalreal-timeimagingstereomatcher.OpticalEngineering,23(5):536--545,1984.

No context found.

K. Nishihara. Practical real-time imaging stereo matcher. Optical Engineering, 23#5#:536#545, 1984.

No context found.

H. K. Nishihara. Practical real-time imaging stereo matcher. Optical Engineering, 23(5):536--545, September, October 1984.

No context found.

H.K. Nishihara, "Practical Real-Time Imaging Stereo Matcher", Optical Engineering, vol. 23, pp. 536-545, Sep. 1984.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC