From nospam at hotmail.com Wed Jun 2 17:36:57 2004
From: nospam at hotmail.com (Skybuck Flying)
Date: Mon Jan 9 13:41:14 2006
Subject: Line segment with grid intersection test ?
Message-ID: <001501c448af$118df850$395d79d9@cp250405a>
Hi,
Problem description:
A line segment intersects with a grid. The cells in the grid have the same
dimensions.
The objective is to mark all cells that are intersected by the line segment.
It seems like a general problem which could be solved by a general
algorithm.
I am looking for a description of an algorithm that will work in 2D.
A very simple solution could be to just do a line segment with rectangle
intersection test for every cell.
Though this doesn't seem efficient.
A more efficient way could be to move along the line segment from cell
boundary to cell boundary.
Any hints, links, etc are appreciated ;)
Bye,
Skybuck.
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From nospam at hotmail.com Wed Jun 2 18:26:41 2004
From: nospam at hotmail.com (Skybuck Flying)
Date: Mon Jan 9 13:41:14 2006
Subject: Line segment with grid intersection test ?
Message-ID: <002601c448b6$03896440$395d79d9@cp250405a>
Actually.
I am looking for a "geometry" solution.
Not a graphical solution.
Graphical solutions can get away with some inprecision.
I need the result to be exact/precise so that it can be used for further
calculations etc.
Usenet does not have a geometry newsgroup so I posted it to graphics since
it seems the next best thing and game development.
I also CC-ed the computional mailing list.
Usenet could use a computational geometry group ! ;)
Skybuck.
"Hans-Bernhard Broeker" wrote in message
news:2i67itFiutgnU1@uni-berlin.de...
> [F'up2 reduced to a single group --- should have been done by OP
> already.]
>
> In comp.graphics.algorithms Skybuck Flying wrote:
>
> > A line segment intersects with a grid. The cells in the grid have the
same
> > dimensions.
>
> > The objective is to mark all cells that are intersected by the line
segment.
>
> > It seems like a general problem which could be solved by a general
> > algorithm.
>
> You're right, and it is. Such algorithms go by the name "scan
> conversion", and include classics like Bresenham's integer-only scan
> conversion. The vital clue is that your regular grid of cells is
> essentially equivalent to a pixelated screen, so what you're doing is
> equivalent to drawing a line segment on such a device.
>
> The only major difference between your task and plain vanilla line
> segment scan conversion is that you seem to require end points at
> arbitrary positions inside a cell/pixel. I.e. you're looking for a
> sub-pixel resolution in scan conversion.
>
> The best choice really is to walk along the line and determine for
> each cell you're in, whether the edge leaves it through a horizontal
> or vertical edge, or exactly at a corner, or ends inside that cell.
> The test to carry out is whether the eligible corner is to the right
> or to the left of the line.
>
> --
> Hans-Bernhard Broeker (broeker@physik.rwth-aachen.de)
> Even if all the snow were burnt, ashes would remain.
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From david.strip at kodak.com Wed Jun 2 17:41:49 2004
From: david.strip at kodak.com (David Strip)
Date: Mon Jan 9 13:41:14 2006
Subject: Line segment with grid intersection test ?
References: <001501c448af$118df850$395d79d9@cp250405a>
Message-ID: <40BE57AD.D5515BC8@kodak.com>
Skybuck Flying wrote:
> Hi,
>
> Problem description:
>
> A line segment intersects with a grid. The cells in the grid have the same
> dimensions.
>
> The objective is to mark all cells that are intersected by the line segment.
>
Look at Bresemham's line drawing algorithm. It does almost what you want. The
Bresemham algorithm would identify the vertices you come closest to. You can
change the marking step to utilized a slightly different decision that will mark
the appropriate grid cells. (At first I thought you might be able to take a dual
grid and use the straight bresenham algorithm on a line plotted in the dual
space, but that doesn't work).
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From ah at cs.unh.edu Wed Jun 2 19:41:39 2004
From: ah at cs.unh.edu (Alejo Hausner)
Date: Mon Jan 9 13:41:14 2006
Subject: Line segment with grid intersection test ?
In-Reply-To: <001501c448af$118df850$395d79d9@cp250405a>
References: <001501c448af$118df850$395d79d9@cp250405a>
Message-ID:
On Wed, 2 Jun 2004, Skybuck Flying wrote:
> A line segment intersects with a grid. The cells in the grid have the same
> dimensions.
>
> The objective is to mark all cells that are intersected by the line segment.
>
> I am looking for a description of an algorithm that will work in 2D.
(I didn't invent the stuff below)
The problem arises in acceleration of ray tracing.
A simple solution will treat the line segment as a
ray, and will march accross the grid. It must keep
track of the distance t along the ray, where the ray
enters the current grid cell (i,j).
1. Find which side of the grid the ray hits first,
determine the initial t, and identify the initial
grid cell (i,j).
2. Consider the two lines on the "other" side of cell
(i,j) : the ray intersects the x-parallel line at
tx, and the y-parallel line at ty.
3. The smaller of tx and ty determines the next cell
visited:
if (tx < ty) {
t = tx;
i++;
}
else {
t = ty;
j++;
}
4. Repeat until ray exits whole grid.
Note: for step 2, you can pre-compute a decision
flag which can be used in the loop to quickly decide
which exit lines you should test.
The algorithm extends very naturally to 3 and higher
dimensions.
------------------------------+-------------------------------
Alejo Hausner (ah@cs.unh.edu) | Mailing address:
phone: (603) 862-1237 | Dept. of Computer Science
fax: (603) 862-3493 | 131 Main St.
Office: Nesmith 303 | Durham, NH 03824
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From nospam at hotmail.com Fri Jun 4 15:35:57 2004
From: nospam at hotmail.com (Skybuck Flying)
Date: Mon Jan 9 13:41:14 2006
Subject: rectangulizing an area
Message-ID: <002e01c44a30$7f838d70$395d79d9@cp250405a>
Hi,
Is there any known algorithm for rectangulizing an area ?
I think I just invented a new word: Rectangulizing :)
Rectangulizing is like Triangulizing; Splitting a rectangle into multiple
rectangles.
I am looking for an algorithm that can split an area into multiple
rectangles.
The new rectangles could have the same height, width ratio as the original
rectangle.
That would be the best case.
But the ratio could also be a bit more flexible or even given.
All new rectangles should have the same dimensions compared to each other.
Ofcourse the new rectangles should cover the original area completely: no
more, no less ;)
( But... for my purposes... the algorithm could get away with a little bit
of extending the area ;) )
The algorithm may not produce more rectangles than a given number of maximum
rectangles.
It's ok if it procedures less rectangles than the maximum, but should try to
create as much as possible up to the maximum ;)
( I haven't given it much though yet, so any links, hints or info in case
such an algorithm already exists are welcome :D )
Bye,
Skybuck.
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From Szoraster at lgc.com Thu Jun 3 13:04:31 2004
From: Szoraster at lgc.com (Steve Zoraster)
Date: Mon Jan 9 13:41:14 2006
Subject: Line segment with grid intersection test ?
Message-ID: <5BAA1AB0910F3A448793DD9A5C4C4E1701ABCD7C@lgchexch009.ad.lgc.com>
Normalize everything so that the origin of the
grid is (0,0). And cell size 1x1.
Divide the line segment into subsegments of length
say 3/4. For each such subsegment determine whether
one end x coordinate is (n).kkkk and the other, for
example (n+1).kkkk. That means a grid column crossing
and even tells you which column. Compute and store
exact column intersection point. That tells you which row.
Do same for y-coordinate. Continue until run out of
all segments.....
Steven Zoraster
Normalize everything so that the origin of the
grid is (0,0). And cell size 1x1.
Divide the line segment into subsegments of length
say 3/4. For each such segment determine whether
one end x coordinate is (n).kkkk and the other, for
example (n+1).kkkk. That means a grid column crossing
and even tells you which column. Compute and store
exact intersection point. That tells you which row.
Do same for y-coordinate. Continue until run out of
all segments.....
Steven Zoraster
Normalize everything so that the origin of the
grid is (0,0). And cell size 1x1.
Divide the line segment into subsegments of length
say 3/4. For each such segment determine whether
one end x coordinate is (n).kkkk and the other, for
example (n+1).kkkk. That means a grid column crossing
and even tells you which column. Compute and store
exact intersection point. That tells you which row.
Do same for y-coordinate. Continue until run out of
all segments.....
Steven Zoraster
> -----Original Message-----
> From: Skybuck Flying [mailto:nospam@hotmail.com]
> Sent: Wednesday, June 02, 2004 9:37 AM
> To: compgeom-discuss@research.bell-labs.com
> Subject: Line segment with grid intersection test ?
>
>
> Hi,
>
> Problem description:
>
> A line segment intersects with a grid. The cells in the grid
> have the same
> dimensions.
>
> The objective is to mark all cells that are intersected by
> the line segment.
>
> It seems like a general problem which could be solved by a general
> algorithm.
>
> I am looking for a description of an algorithm that will work in 2D.
>
> A very simple solution could be to just do a line segment
> with rectangle
> intersection test for every cell.
>
> Though this doesn't seem efficient.
>
> A more efficient way could be to move along the line segment from cell
> boundary to cell boundary.
>
> Any hints, links, etc are appreciated ;)
>
> Bye,
> Skybuck.
>
>
>
>
> -------------
> The compgeom mailing lists: see
> http://netlib.bell-labs.com/netlib/compgeom/readme.html
> or send mail to compgeom-request@research.bell-labs.com with the line:
> send readme
> Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
> >
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From bender at cs.sunysb.edu Thu Jun 3 19:50:18 2004
From: bender at cs.sunysb.edu (Michael Bender)
Date: Mon Jan 9 13:41:14 2006
Subject: SPAA 2004 Call for Participation
Message-ID:
Call for Participation
==========================================================================
Sixteenth ACM Symposium on Parallelism in Algorithms and Architectures
SPAA '04
June 27-30, 2004
Barcelona, Spain
==========================================================================
This symposium was called the ACM Symposium on Parallel Algorithms and
Architectures from 1989 to 2002. The new name to reflects the expanded
scope of the conference as detailed below.
This year, in a tradition starting in 2001, SPAA defines "parallel" very
broadly to encompass any computational "device" that can perform multiple
operations or tasks simultaneously. As a consequence, SPAA 2004 covers all
aspects of parallelism. This includes traditional parallel and distributed
algorithms and architectures, plus new aspects including the internet, the
web, sensor networks, quantum and DNA computing, etc. For more
information, see the call for papers.
==========================================================================
Venue
The conference will be held at the Technical University of Catalonia.
See http://www.spaa-conference.org/ for more information.
==========================================================================
Registration
The deadline for early registration is June 7, 2004.
See http://www.spaa-conference.org/ for registration information.
==========================================================================
Technical Program
June 27, 2004, Sunday
6pm - TBD Registration
7pm - TBD Reception at Catedra Gaudi (Campus Nord - UPC)
June 28, 2004, Monday
TBD Continental Breakfast
9:00am - 9:10am Welcome
9:10am - 10:25am Session 1: Routing I
9:10am - 9:35am On Delivery Times in Packet Networks under Adversarial Traffic
Adi Rosen and Michael S. Tsirkin
9:35am - 10:00am Adaptive Channel Queue Routing on k-ary n-cubes
Arjun Singh, William J. Dally, Amit Gupta, and Brian Towles
10:00am - 10:25am Compact Name-Independent Routing with Minimum Stretch
Ittai Abraham, Cyril Gavoille, Dahlia Malkhi, Noam Nisan, and Mikkel Thorup
10:25am - 10:50am Break
10:50am - 12:30pm Session 2: Peer-To-Peer Systems
10:50am - 11:15am Object Location in Realistic Networks
Kirsten Hildrum, Robert Krauthgamer, and John Kubiatowicz
11:15am - 11:40am Simple Efficient Load Balancing Algorithms for Peer-to-Peer Systems
David R. Karger and Matthias Ruhl
11:40am - 12:05am Consistent, Order-Preserving Data Management in Distributed Storage Systems
Baruch Awerbuch and Christian Scheideler
12:05am - 12:30pm Geometric Generalizations of the Power of Two Choices
John W. Byers, Jeffrey Considine, and Michael Mitzenmacher
12:30pm - 2:30pm Lunch
2:30pm - 3:45pm Session 3: Caching I
2:30pm - 2:55pm Fighting Against Two Adversaries: Page Migration in Dynamic Networks
Marcin Bienkowski, Miroslaw Korzeniowski, and Friedhelm Meyer auf der Heide
2:55pm - 3:20pm Online Hierarchical Cooperative Caching
Xiaozhou Li, C. Greg Plaxton, Mitul Tiwari, and Arun Venkataramani
3:20pm - 3:45pm New Results on Web Caching with Request Reordering
Susanne Albers
3:45pm - 4:15pm Break
4:15pm - 5:30pm Session 4: Routing II
4:15pm - 4:40pm Packet-Mode Policies for Input-Queued Switches
Dan Guez, Alex Kesselman, and Adi Rosen
4:40pm - 5:05pm Parallelism versus Memory Allocation in Pipelined Router Forwarding Engines
Fan Chung, Ronald Graham, and George Varghese
5:05pm - 5:30pm Lower Bounds for Approximating Sparse Graphs and M-Matrices
Gary L. Miller and Peter C. Richter
Evening, time TBD Business Meeting
June 29, 2004, Tuesday
TBD Continental Breakfast
9:10am - 10:50am Session 5: Algorithms
9:10am - 9:35am Balanced Graph Partitioning
Konstantin Andreev and Harald Raecke
9:35am - 10:00am Bi-criteria Algorithm for Scheduling Jobs on Cluster Platforms
Pierre-Francois Dutot, Lionel Eyraud, Gregory Mounie, and Denis Trystram
10:00am - 10:25am On-the-Fly Maintenance of Series-Parallel Relationships in Fork-Join
Multithreaded Programs
Michael A. Bender, Jeremy T. Fineman, Seth Gilbert, and Charles E. Leiserson
10:25am - 10:50am An NC Algorithm for Finding a Maximal Acyclic Set in a Graph
Aaron Windsor
10:50am - 11:15am Break
11:15am - 12:30pm Session 6: Networks I
11:15am - 11:40am Scheduling Against an Adversarial Network
Stefano Leonardi, Alberto Marchetti-Spaccamela, and Friedhelm Meyer auf der Heide
11:40am - 12:05am On Achieving Optimized Capacity Utilization in Application Overlay Networks with
Multiple Competing Sessions
Yi Cui, Baochun Li, and Klara Nahrstedt
12:05am - 12:30am Pagoda: A Dynamic Overlay Network for Routing, Data Management, and Multicasting
Ankur Bhargava, Kishore Kothapalli, Chris Riley, Christian Scheideler, and Mark Thober
12:30pm - 2:30pm Lunch
2:30pm - 3:45pm Session 7: Algorithmic Game Theory
2:30pm - 2:55pm Sharing the Cost of Multicast Transmissions in Wireless Network
V. Bilo, C. Di Francescomarino, M. Flammini, and G. Melideo
2:55pm - 3:20pm Selfish Load Balancing and Atomic Congestion Games
Subhash Suri, Csaba D. Toth, and Yunhong Zhou
3:20pm - 3:45pm How to Route and Tax Selfish Unsplittable Traffic
Vincenzo Auletta, Roberto De Prisco, Paolo Penna, and Pino Persiano
Evening Outing and Dinner
June 30, 2004, Wednesday
TBD Continental Breakfast
9:00am - 10:15am Session 8: Shared Memory and Architecture
9:00am - 9:25am A Scalable Lock-free Stack Algorithm
Danny Hendler, Nir Shavit, and Lena Yerushalmi
9:25am - 9:50am DCAS is not a Silver Bullet for Nonblocking Algorithm Design
Simon Doherty, David L. Detlefs, Lindsay Groves, Christine H. Flood,
Victor Luchangco, Paul A. Martin, Mark Moir, Nir Shavit, and Guy L. Steele, Jr.
9:50am - 10:15am Efficient Orchtestration of Sub-Word Parallelism in Media Processors
John Oliver, Venkatesh Akella, and Frederic T. Chong
10:15am - 10:30am Break
10:30am - 11:45pm Session 9: Caching II
10:30am - 10:55am Effectively Sharing a Cache Among Threads
Guy E. Blelloch and Phillip B. Gibbons
10:55am - 11:20am Cache-Oblivious Shortest Paths in Graphs Using Buffer Heap
Rezaul Alam Chowdhury and Vijaya Ramachandran
11:20am - 11:45am Online Algorithms for Prefetching and Caching on Parallel Disks
Rahul Shah, Peter Varman, and Jeffrey Scott Vitter
11:45am - 12:00am Break
12:00pm - 12:50pm SPAA Revue
12:00pm - 12:10pm Improved Combination of Online Algorithms for Acceptance and Rejection
David P. Bunde and Yishay Mansour
12:10pm - 12:20pm Time Complexity of Practical Parallel Steiner Point Insertion Algorithms
Dan A. Spielman, Shang-hua Teng, and Alper Ungor
12:20pm - 12:30pm The Inherent Queuing Delay of Parallel Packet Switches
Hagit Attiya and David Hay
12:30pm - 12:40pm Efficient Search in Unstructured Peer-to-Peer Networks
Vicent Cholvi, Pascal Felber, and Ernst Biersack
12:40pm - 12:50pm The Potential in Energy Efficiency of a Speculative Chip-Multiprocessor
Yuu Tanaka and Toshinori Sato and Takenori Koushiro
12:50pm - 2:45pm Lunch
2:45pm - 4:00pm Session 10: Networks II
2:45pm - 3:10pm Online Algorithms for Network Design
Adam Meyerson
3:10pm - 3:35pm Expansion Properties of (Secure) Wireless Networks
Alessandro Panconesi and Jaikumar Radhakrishnan
3:35pm - 4:00pm The Effect of Random Faults on Network Expansion
Amitabha Bagchi, Ankur Bhargava, Amitabh Chaudhary, David Eppstein,
and Christian Scheideler
4:00pm - 4:15pm Break
4:15pm - 5:30pm Session 11: Distributed Computation
4:15pm - 4:40pm Dynamic Analysis of the Arrow Distributed Protocol
Fabian Kuhn and Roger Wattenhofer
4:40pm - 5:05pm Optimal Early Stopping Uniform Consensus in Synchronous Systems
with Process Omission Failures
Philippe Raipin Parvedy and Michel Raynal
5:05pm - 5:30pm Writing-All Deterministically and Optimally Using a Non-Trivial
Number of Asynchronous Processors
Dariusz Kowalski and Alex Shvartsman
5:30pm END
==========================================================================
Program Chair
Micah Adler, U. Massachusetts
Program Committee
Micah Adler, U. Massachusetts
John Byers, Boston U.
Tom Cormen, Dartmouth College
Bruce Hendrickson, Sandia National Laboratories
Maurice Herlihy, Brown U.
Christos Kaklamanis, U. Patras
Christian Lengauer, U. Passau
Geppino Pucci, U. Padova
Satish Rao, U.C. Berkeley
Yves Robert, ENS Lyon
Peter Sanders, MPI Saarbrucken
Daniel Sorin, Duke U.
Aravind Srinivasan, U. Maryland
Berthold Vocking, U. Dortmund
SPAA Local Arrangements Chair
Eulalia Barriere, Technical U. of Catalonia
SPAA General Chair
Phil Gibbons, Intel Research
SPAA Secretary
Cynthia A. Phillips, Sandia National Laboratories
SPAA Treasurer
Rajmohan Rajaraman, Northeastern U.
Publicity Chair
Michael Bender, SUNY Stony Brook
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From Sylvain.Petitjean at loria.fr Wed Jun 16 12:07:42 2004
From: Sylvain.Petitjean at loria.fr (Sylvain Petitjean)
Date: Mon Jan 9 13:41:14 2006
Subject: QI 1.0 released - Intersection of quadrics
Message-ID: <40D00DDE.8020701@loria.fr>
We are pleased to announce the first release of QI, our software for
computing near-optimal parameterizations of the intersection of implicit
quadrics. It is available from the QI web site
http://www.loria.fr/isa/qi
QI has the following features:
* it computes an exact parameterization of the intersection of two
quadrics with integer coefficients of arbitrary size;
* it correctly identifies, separates and parameterizes all the
connected components of the intersection and gives all the
relevant topological information;
* the parameterization is rational when one exists; otherwise the
intersection is a smooth quartic and the parameterization involves
the square root of a polynomial;
* the parameterization is either optimal in the degree of the
extension of the integers on which its coefficients are defined
or, in a small number of well-identified cases, involves one extra
possibly unnecessary square root;
* it is fast and efficient and can routinely compute
parameterizations of the intersection of quadrics with input
coefficients having ten digits in less than 50 milliseconds on a
mainstream PC.
QI is available free of charge for non-commercial use. It was developed
at the LORIA laboratory, Nancy, France. It is based on research work led
by Laurent Dupont, Daniel Lazard, Sylvain Lazard and Sylvain Petitjean.
Comments and questions should be addressed to Sylvain.Petitjean@loria.fr
and Sylvain.Lazard@loria.fr.
--
- Sylvain Petitjean and Sylvain Lazard
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://compgeom.poly.edu/pipermail/compgeom-announce/attachments/20040616/3ec3eb5d/attachment.htm
From Sylvain.Petitjean at loria.fr Wed Jun 16 19:31:46 2004
From: Sylvain.Petitjean at loria.fr (Sylvain Petitjean)
Date: Mon Jan 9 13:41:14 2006
Subject: PhD fellowship - Effective computational geometry
Message-ID: <40D075F2.30900@loria.fr>
(Please pass to interested parties.)
Applications are invited for a PhD fellowship in the field of effective
computational geometry and geometric computing. The PhD topic is on
robust and
efficient calculation of arrangements of quadrics (degree 2 surfaces) --
see below
for more details.
Research will take place at the LORIA laboratory, Nancy, France, in the ISA
project-team (Models, algorithms and geometry for computer graphics and
artificial
vision). See the web pages for further info: http://www.loria.fr and
http://www.loria.fr/isa
The PhD grant is slightly above 1300 euros per month over 36 months. The
thesis is
scheduled to start in October 2004.
The ideal candidate will have working knowledge of computational
geometry and a
solid background in mathematics.
The exact conditions to be allowed to apply for this fellowship are
stated on the
following page (in French):
http://dr.education.fr/Alloc_doc/alloc_2.html
Roughly: to be under 25, to have or be about to have a Master's degree,
to be a
citizen of a member state of the European Union. There are also special
conditions
for citizens of Bulgaria, Russia, Iceland, Norway, Roumania, Switzerland and
Turkey, so feel free to send e-mail to enquire.
To apply, please send a fairly complete vitae by e-mail to
Sylvain.Petitjean@loria.fr
The schedule is very tight: deadline for application is set to July 5th,
2004.
Contact person:
Sylvain Petitjean
LORIA
Campus scientifique
BP 239
54506 Vandoeuvre cedex
FRANCE
-----------------
Efficient and Robust Computation of Arrangements of Quadrics
------------------------------------------------------------
The two most widely used types of object representation in solid
modeling are
constructive solid geometry (CSG) and boundary representation (BRep). Both
representations having their own respective advantages, solid modeling
kernels
often need an efficient and reliable way to convert one type of
representation
into the other. CSG-to-BRep conversion is a well understood problem, but
past
approaches have often put more emphasis on efficiency than on robustness and
accuracy. If only finite-precision arithmetic is assumed, the topological
consistency of the compute BRep can easily be jeopardized by small
amounts of
error introduced in the data. For many applications in design and automated
manufacturing, this may be unacceptable.
Designing reliable and robust algorithms is currently a major interest
of the
computational geometry and geometric computing research communities. A
number of
successful approaches have been proposed for the robust and accurate
CSG-to-BRep
conversion of polyhedral models. Computing the topological structure of
a BRep
involves accurately evaluating signs of arithmetic expressions, which can be
achieved for piecewise-linear models assuming the necessary bit length
is allowed
for number representation.
By contrast, there has been much less work on robust CSG-to-BRep conversion
algorithms for curved primitives (but see [1,2,3]). A major reason is
that outside
the linear realm exact arithmetic computations require algebraic numbers
which
cannot in general be represented explicitly with a finite number of bits. In
addition, computation with algebraic numbers is extremely slow.
In the algebraic domain, the most simple surfaces, outside piecewise-linear
meshes, are degree 2 surfaces, i.e. quadrics. Quadrics represent a
fairly good
compromise between complexity, flexibility and modeling power. With
respect to the
CSG-to-BRep conversion problem, and more generally to the computation of
arrangements of quadrics, they also have undeniable advantages. Indeed, the
quadratic nature of their defining equations permits an explicit
representation of
intersection curves. Thus, it is theoretically possible to compute a fully
parametric representation of the boundary of second-order CSG solids.
The goal of this PhD is to make strides towards this goal. Recently, we have
proposed and implemented a robust algorithm for the near-optimal
parameterization
of the intersection of two quadrics [4,5]. The complexity of the output
is minimal
in terms of the number and depth of the radicals involved. Using this
algorithm
(and the accompanying theoretical results) as a building block, the doctoral
candidate will propose solutions for the robust and efficient computation of
arrangements of quadrics. Depending on his/her background, he/she will
focus more
specifically on one or several of the many different aspects of the problem:
handling degenerate cases, mixing finite-precision and exact arithmetic,
ensuring the geometrical-topological consistency, using filters,
designing proper
geometric predicates, managing complexity, solving polynomial equations with
algebraic coefficients, sorting algebraic points along a curve, ...
[1] J. Keyser, S. Krishnan and D. Manocha, Efficient and Accurate BRep
Generation
of Low Degree Sculptured Solids Using Exact Arithmetic, Computer Aided
Geometric
Design, 16 (9), 1999.
[2] E. Sch?mer and N. Wolpert, An Exact and Efficient Approach for
Computing a
Cell in an Arrangement of Quadrics, Computational Geometry: Theory and
Applications, 2004.
[3] B. Mourrain, J.-P. T?court and Monique Teillaud, Predicates for the
Sweeping
of an Arrangement of Quadrics in 3D, Computational Geometry: Theory and
Applications, 2004.
[4] L. Dupont, D. Lazard, S. Lazard and S. Petitjean, Near-Optimal
Parameterization of the Intersection of Quadrics, Proc. of the ACM
Symposium on
Computational Geometry, 2003.
[5] S. Lazard, L. M. Pe?aranda and S. Petitjean, Intersecting Quadrics: An
Efficient and Exact Implementation, Proc. of the ACM Symposium on
Computational
Geometry, 2004
--
- Sylvain
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From vsr at ccad.uiowa.edu Sat Jun 5 11:32:31 2004
From: vsr at ccad.uiowa.edu (Virtual Soldier Research)
Date: Mon Jan 9 13:41:14 2006
Subject: A sneak preview of the Digital Human Environment SANTOS at the
SAE conference
Message-ID: <6.0.0.22.2.20040605102758.0237d4e0@renoir.ccad.uiowa.edu>
A sneak preview of the Digital Human Environment SANTOS at the SAE conference
Form Immediate Release
Iowa City, IA--June 4, 2004: The Virtual Soldier Research (VSR) program at
the University of Iowa announced today that members of the team will be
providing a sneak preview of the Santos digital human environment during
the annual meeting of the Society of Automotive Engineers (SAE) Digital
Human Modeling Conference in Oakland, MI, June 14-16, 2004. Santos has
deformable skin, contracting muscles, is anatomically correct and lives in
a real-time interactive immersive virtual reality environment. The
environment will be used for design, manufacturing, sports, training,
maintenance, and communication of industry and military products. A date
for releasing Santos has not been specified. For more information about
Santos please visit the following web site:
http://www.engineering.uiowa.edu/~amalek/VSR
About VSR
VSR is an independent research group within the Center for Computer Aided
Design at The University of Iowa. Funded primarily by the US Army TACOM,
VSR conducts basic and applied research for creating new technologies
dealing with digital human modeling and simulation. Researchers (faculty,
staff, scientists, engineers, clinical researchers, and graduate students)
from various fields that include engineering, gaming, psychology,
biomechanics, human factors, computers, optimization, and industrial design
have come together to create this new technology. The objective is to
create human-like life on the computer, virtual humans/soldiers that can
walk, behave, and talk like we do, yet are able to answer questions in the
virtual world autonomously. Rather than building a vehicle, a tank, or a
weapon system, we allow the virtual soldier to experience the product in
the virtual world, thus providing feedback without building a prototype.
Virtual Soldier Research (VSR) Program
Center for Computer Aided Design
THE UNIVERSITY OF IOWA
Iowa City, IA. 52242
Tel: (319) 335-5722
Fax: (319) 384-0542
vsr@ccad.uiowa.edu
www.engineering.uiowa.edu/~amalek/VSR
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://compgeom.poly.edu/pipermail/compgeom-announce/attachments/20040605/b45e6aa8/attachment.htm
From casa2004 at MIRALAB.UNIGE.CH Thu Jun 10 14:34:27 2004
From: casa2004 at MIRALAB.UNIGE.CH (Casa 2004)
Date: Mon Jan 9 13:41:14 2006
Subject: [CASA2004] Call for Participation: CASA 2004 Conference
Message-ID:
[Apologies if you receive this CFP more than once]
CALL FOR PARTICIPATION TO CASA 2004 AT GENEVA, SWITZERLAND
Dates and Place: July 7-9, 2004 at Geneva, Switzerland Conference
Information: http://casa2004.miralab.unige.ch/
The COMPUTER GRAPHICS SOCIETY (CGS) is pleased to announce
the COMPUTER ANIMATION and SOCIAL AGENTS 2003 Conference
to be held at University of Geneva, Switzerland. MIRALab
(http://www.miralab.ch)
will organize this 17th annual conference on Computer Animation and Social
Agents
(CASA2004) with the support of IFIP WG5.10 (Computer Graphics and Virtual
Worlds).
The CASA conference has been the most exciting place to meet researchers and
discuss advancement on Computer Animation and Social Agent technologies for
decades.
This year, total 77 papers, which went through very high competition on
reviewing process,
will be presented along with three tutorials and a panel discussion in the
area of (but not limited to);
- Animation with Emotion
- Autonomous Agents
- Social Agents
- Behavioral Animation
- Natural Phenomena
- Crowd Modeling
- Animating Deformable Objects
- Motion Control
- Virtual Reality
- HCI
With all paper presentations, tutorials and panel discussions, CASA 2004
will elaborate
discussions on the past, current and future of the computer animation
technologies and
advances in the social agents.
For further information on program, venue and registration, please visit at;
http://casa2004.miralab.unige.ch/
Hope to meet you at CASA 2004, Geneva.
Sincerely yours,
CASA 2004 Local Committee
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From fd at dehne.net Fri Jun 11 17:42:26 2004
From: fd at dehne.net (Prof. Frank Dehne (www.dehne.net))
Date: Mon Jan 9 13:41:14 2006
Subject: Computing: the Australasian Theory Symposium (CATS'05)
Message-ID:
CALL FOR PAPERS
Computing: the Australasian Theory Symposium (CATS'05)
CATS'05 will take place as part of Australasian Computer Science Week
(ACSW) at the University of Newcastle 30 January to 3 February. Papers
are invited in all areas of theoretical computer science.
Submissions should be made by 3 September 2004 via the conference web
site: http://www.cs.otago.ac.nz/staffpriv/mike/CATS05/CATS05.html.
Conference participants should register through the ACSW web site:
http://www.cs.newcastle.edu.au/~acsw05.
Papers are invited on all aspects of Theoretical Computer Science.
Some representative, but not exclusive, topics include the following:
logic, reasoning and verification
formal specification techniques and program semantics
formal development methods, program refinement, synthesis and transformation
concurrent, parallel and distributed system theory
algorithm design and data structures
streaming data computation,
computational biology, geometry, and number theory.
complexity and computability
automata, types and category theory
tools for automated reasoning, and program analysis and development
Important Dates:
Friday 3rd September, 2004: Deadline for submissions of full papers
Friday 15th October, 2004: Notification of acceptance for formal submissions
Friday 26th November, 2004: Deadline for informal submissions
Friday 12th November, 2004:
Final versions of accepted formal papers due
Deadline for author registrations
Friday 3rd December, 2004: Notification of acceptance for informal submissions
Sunday 30th January to Thursday 3rd February, 2005: Australasian Computer Science Week, incorporating CATS 2005
CATS'05 Program Committee:
Mike Atkinson (Co-chair), University of Otago, New Zealand.
Frank Dehne (Co-chair), Griffith University, Australia.
James Harland, RMIT University, Melbourne, Australia.
Barry Jay, University of Technology, Sydney, Australia.
Mike Johnson, Macquarie University, Sydney, Australia.
David Wolfram, Infosys Australia.
Rod Downey, Victoria University of Wellington, New Zealand.
Angele Hamel, Wilfrid Laurier University, Canada.
Joerg Sack, Carleton University, Canada.
Roy Dyckhoff, University of St Andrews, UK.
Andrew Rau-Chaplin, Dalhousie University, Canada.
Bakhadyr Khoussainov, University of Auckland, New Zealand.
Bruce Litow, James Cook University.
Rodney Topor, Griffith University.
Andrew Solomon, University of Technology, Sydney.
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.
From qzg00130 at scc.u-tokai.ac.jp Fri Jun 11 17:26:54 2004
From: qzg00130 at scc.u-tokai.ac.jp (Masatsugu Urabe)
Date: Mon Jan 9 13:41:14 2006
Subject: JCDCG2004
Message-ID: <200406111626.IBD49356.2RY400H0@scc.u-tokai.ac.jp>
(my apologies for multiple copies.)
CALL FOR PAPERS
============================================================
Japan Conference on Discrete and Computational Geometry 2004
(JCDCG2004)
Date: October 8-11, 2004
Place: Yoyogi Campus of the Tokai University (Tokyo)
============================================================
The conference is intended to provide a forum for researchers and
R&D people dealing with all aspects of discrete and computational
geometry. JCDCG has been held annually since 1997. In particular,
JCDCG 2004 will be held as a conference in Honor of Janos Pach
on his 50th Year.
The proceedings of the conference will be published as a volume of
Lecture Notes in Computer Science (LNCS, Springer-Verlag).
Important Dates
----------------
Submission deadline for extended abstracts:
The official deadline for submission is August 20, 2004.
Notification of acceptance will be sent to you on September 10, 2004.
For those who wish to receive the notification from us earlier so that
you can start preparing (i.e. getting a visa and other documents),
send your extended abstract by June 30, 2004. In this case,
notification will be sent by July 10, 2004.
Submission deadline for full papers (for Proceedings):
October 11, 2004
Invited Speakers
----------------
Ferran Hurtado (Universitat Politecnics de Catalunya)
Hiro Ito (Kyoto University)
Juri Matousek (Charles University)
Janos Pach (City College and Courant Institute, New York;
Alfred Renyi Institute of Mathematics, Budapest)
Jonathan Shewchuk (University of California at Berkeley)
Endre Szemeredi (Rutgers University)
Geza Toth (Renyi Institute of Mathematics, Budapest)
Godfried Toussaint (McGill University)?
Yinfeng Xu (Xi'an Jiaotong University)
Details will be informed in the Conference Home Page:
http://www.ried.tokai.ac.jp/JCDCG
Conference Office:
Research Institute of Educational Development, Tokai University
2-28-4 Tomigaya, Shibuya-ku,
Tokyo 151-8677, Japan
Phone and Fax: +81-3-3485-4979
E-mail: jcdcg@ried.tokai.ac.jp
We would like to request you to pass on the information about the
conference to your colleagues. It will be our pleasure to have you
and your colleagues could attend the conference.
--
Masatsugu Urabe
qzg00130@scc.u-tokai.ac.jp
Tokai University
3-20-1 ShimizuOrido Shizuoka 424-8610 JAPAN
Tel. +81-543-34-0411 (Ext. 3436, 3116)
Fax. +81-543-34-9837
-------------
The compgeom mailing lists: see
http://netlib.bell-labs.com/netlib/compgeom/readme.html
or send mail to compgeom-request@research.bell-labs.com with the line:
send readme
Now archived at http://www.uiuc.edu/~sariel/CG/compgeom/maillist.html.