Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!usc!apple!agate!ucbvax!ADS.COM!vision From: vision@ADS.COM (Philip Kahn) Newsgroups: comp.ai.vision Subject: (none) Message-ID: <9101300500.AA00888@deimos.ads.com> Date: 30 Jan 91 05:00:25 GMT Sender: daemon@ucbvax.BERKELEY.EDU Distribution: inet Organization: The Internet Lines: 305 Approved: vision-list@ads.com Date: 29 Jan 91 13:44:43-PST From: Vision-List moderator Phil Kahn Errors-to: Vision-List-Request@ADS.COM Reply-to: Vision-List@ADS.COM Subject: VISION-LIST digest 10.5 To: Vision-List@ADS.COM VISION-LIST Digest Tue Jan 29 13:44:43 PDT 91 Volume 10 : Issue 5 - Send submissions to Vision-List@ADS.COM - Send requests for list membership to Vision-List-Request@ADS.COM - Access Vision List Archives via anonymous ftp to ADS.COM Today's Topics: IMDISP, Magellan Images Available Call for Papers: IJIVC Boston SPIE Robots and Computer Vision Conference tracking stereo (summary) ---------------------------------------------------------------------- Date: Fri, 25 Jan 91 7:42:52 PDT From: baalke@mars.Jpl.Nasa.Gov (Ron Baalke) Subject: IMDISP, Magellan Images Available ============ IMDISP 5.6 ============ Version 5.6 of the IMDISP program is now available via anonymous ftp. IMDISP is an interactive image display program with some basic image processing capabilities that runs on an IBM PC computer. It was originally written to be used with the Voyager images on CD-ROM's, and an older version of IMDISP has been distributed with the Voyager CD-ROM's in the past. This version of IMDISP is the most current one, and has been submitted to be included on the GRIPS '90 CD-ROM that is coming out shortly. IMDISP will support various video modes including CGA, EGA, VGA (recommended), and super VGA (highly recommended) for selected graphics boards. The program can display images in three formats: PDS/SFDU, FITS and VICAR. The PDS format was developed by the Planetary Data Systems, and the FITS (Flexible Image Transport) format is typically used for astrononmy images. The VICAR (Video Image Communication and Retrieval) format was developed at the Image Processing Lab at JPL. Peter Yee at Ames Research Center was kind enough to allow his computer site be used for anonymous ftp access. This site can be reached at: ames.arc.nasa.gov (128.102.18.3) The IMDISP program is stored in a file called IMDISP56.ZIP in the pub/SPACE/IMDISP subdirectory. This zip file can be unzipped with the public domain program PKUNZIP. IMDISP56.ZIP has three files in it: IMDISP.EXE - IMDISP program version 5.6 IMDISP.DOC - User's Guide for IMDISP PCDCOMP.EXE - Version 2.0 of the Voyager Image Decompression Program IMDISP.DOC will help in the use of IMDISP, and has some interesting overviews of CD-ROM fundamentals and image processing. The PCDCOMP program is to be used specifically to uncompress the Voyager images stored the CD-ROM into PDS, FITS or VICAR formats; this is the most recent version. Being one of the programmers of IMDISP, I am very interested on how people like the program. If you find any bugs in IMDISP then let me know and I will see that they get fixed. Also, any suggestions or ideas on improving the program will be appreciated. ================= MAGELLAN IMAGES ================= Two raw format Magellan images are also available for anonymous ftp. The two images are stored at the same Ames site listed above, and are in the pub/SPACE/VICAR directory under the names DSNALL.IMG and GOLUBKINA.IMG. The images have been released by the Magellan project at JPL, and are digital versions of images that have been previously released to the press. Both images are 8 bit images in VICAR format with a pixel resolution of 1024x1024, and are 1 megabyte in size. DSNALL.IMG is the "Crater Farm" mosaic which contains 24 orbits of data formed into one large mosaic. The dominating features in this mosaic are three huge craters. GOLUBKINA.IMG is a pseudo-oblique view of a crater from the very first strip of imagery received from Magellan. Note that both these images can be displayed with the IMDISP program. One thing to keep in mind when viewing these images, is that they are radar images and are to be interpreted differently from what we are normally used to seeing. Generally with radar images, dark areas indicate smooth surfaces, and the lighter areas indicate the surface is rough. Additional Magellan images will be added as they become available. ___ _____ ___ /_ /| /____/ \ /_ /| | | | | __ \ /| | | | Ron Baalke | baalke@mars.jpl.nasa.gov ___| | | | |__) |/ | | |___ Jet Propulsion Lab | baalke@jems.jpl.nasa.gov /___| | | | ___/ | |/__ /| M/S 301-355 | |_____|/ |_|/ |_____|/ Pasadena, CA 91109 | ------------------------------ Date: Sun, 27 Jan 91 12:34:35 -0500 From: Baba Vemuri Subject: Call for Papers: IJIVC CALL FOR PAPERS International Journal of Image and Vision Computing announces a special issue on Range Image Understanding Guest Editors: Dr. J. K. Aggarwal, University of Texas, Austin, Texas, USA and Dr. B. C. Vemuri, Dept. of CIS, University of Florida Gainesville, Florida, USA Deadline for submitted papers: January 31, 1991; Publication: June 1991 The understanding of range images is a difficult yet an important area of research in computer vision. Advances in sensor technology for direct 3D sensing in conjunction with progress in passive range sensing methods, have spurred a tremendous growth of research efforts in 3D computer vision. The advent of faster hardware in the form of multi-processing systems has alleviated the burden of processing range data and thus provided further impetus for research in this field of computer vision. The problems of 3D representation, and matching in static and dynamic real-world scenes are formidable research tasks currently being addressed by many researchers. This special issue will gather significant research results on 3D object representation, matching for pose determination and object recognition in static and dynamic environments. Papers describing novel contributions in all aspects of range image understanding are invited, with particular emphasis on: Novel 3D object representation that support range image understanding Matching for pose determination as well as object recognition Representation and indexing of large libraries of 3D objects Applications (eg. parts picking, cartography, navigation etc.) Image and Vision Computing publishes fully refreed technical papers. All submissions will be refreed by two acknowledged authorities. Authors should consult the `Notes for Authors,' published in the journal for detailed preparation of their manuscript (also available on request from the editors). Four single sided double spaced copies of the manuscript should be sent by January 31, 1991, to: Dr. B. C. Vemuri Dept. of Computer and Information Sciences University of Florida Gainesville, Florida 32611 USA ------------------------------ Date: Mon, 28 Jan 91 11:25:38 EST From: Rosalind W. Picard Subject: Boston SPIE Robots and Computer Vision Conference ANNOUNCEMENT and CALL FOR PAPERS Intelligent Robots and Computer Vision X: Algorithms and Techniques Part of SPIE's Technical Symposium on Advances in Intelligent Systems November 10-15, 1991 Boston Mariott Copley Place Boston, Massachusetts USA Chair: David Casasent Carnegie Mellon University Co-Chair: Ernie Hall University of Cincinnati This year's conference will focus on new algorithms and techniques for intelligent robots and computer vision. Papers are solicited for several topic areas including: - pattern recognition and image processing - image understanding and scene analysis - color image processing, multi-sensor processing - 3-D vision: modeling and representation - object modeling and recognition - neural networks, model-based processors - fuzzy logic in intelligent systems and computer vision - biological basis for the design of sensors in computer vision - intelligent manipulators Abstract Due Date: April 1, 1990 Manuscript Due Date: October 14, 1990 Information: SPIE Technical Program Committee/Boston '91 P.O. Box 10 Bellingham, WA 98227-0010 USA Or e-mail to: turk@media-lab.media.mit.edu or marlene@gauss.ece.cmu.edu ------------------------------ Date: Mon, 28 Jan 91 9:26:00 MET DST From: bellutta@irst.it (Paolo Bellutta) Subject: tracking stereo (summary) Organization: I.R.S.T. 38050 POVO (TRENTO) ITALY Here is the list of replies I had from my previous posting. I wish to thank all of the contributors for their promt reply. Paolo Bellutta I.R.S.T. vox: +39 461 814417 loc. Pante' di Povo fax: +39 461 810851 38050 POVO (TN) e-mail: bellutta@irst.uucp ITALY bellutta%irst@uunet.uu.net ********** SUMMARY of REPLIES *************** Date: Sat, 26 Jan 91 14:00:53 PST From: chandy@hotspur.usc.edu (Sundaresan Chandrashekhar) Subject: Your posting on comp.ai.vision I believe something similar to what you suggest has been tried at Carnegie-Mellon by Takeo Kanade. Sorry, I don't have the references for this. From: Xie Ming Subject: Re: tracking stereo Date: Fri, 25 Jan 91 15:26:26 +0100 Hello, These is a Ph.D thesis intitled "Contribution to the dynamic vision: reconstruction of a 3D polyhedral scene by a moving camera" (1989). You could ask one copy of this thesis from: Publication Service INRIA-IRISA campus de Beaulieu 35042 RENNES Cedex France (Standard Phone: (33) 99-36-20-00 ) From: na@bora.inria.fr (Nicholas Ayache) Subject: tracking stereo Date: Fri, 25 Jan 91 09:51:06 +0100 dear Paolo Belluta, you can have a look at our paper entitled 3D structure from a monocular sequence of images by Jezouin and Ayache which produces a 3D representation from tracked tokens (points and lines) in a time sequence of monocular images. Results are very accurate, as you will see. I can send you the corresponding Inria technical report which is more detailed. Sincerely, Nicholas Ayache. Date: Fri, 25 Jan 91 01:02:21 EST From: ks5k@abraxas.cs.Virginia.EDU Subject: tracking stereo Paolo, I have a reference here that may help you out a bit: "One-Eyed Stereo: A General Approach to Modeling 3-D Scene Geometry," Thomas M. Strat, Martin A. Fischler, IEEE PAMI, 8(6): 730-741, Nov 1986. also it is in: _Readings in COmputer Vision, Issues, Problems, Principles, and Paradigms_ Edited by Martin Fischler and Oscar Firschein, 1987, Morgan-Kaufman, ISBN- 0-934613-33-8. It is a 4cm thick volume of useful vision papers. I recommend it highly if you don't have it. It is basically coincidence that I am referencing Fischler's paper, as he only included one or two of his in this approximately 100 paper set. If I remember right this paper basically breaks down to a shape-from-X survey and then presents some experimental results. I don't remember if there is explicit mention of tracking. For tracking you might find things under "image flow" (there are a few papers in the Fischler volume). Do you have the rosenfeld bibliographies there? If you need copies of papers you can not get, I could send them to you if you'd like. Kristian ------------------------------ End of VISION-LIST digest 10.5 ************************