Hi Tharindu,<br><br>I'm also sending this email to the vtk developers list in case other people have insight into your work.<br><br>I'm not sure I completely understand the problem. Can you just replace the implementation for FindCell() with a better implementation without affecting other class methods? If you can do that, I'd just worry about that first and then do some profiling/performance testing to see where it's at. After that it may be worthwhile to see if some of the other class methods can be improved too. If a decent amount of changes have to be made in order to implement that algorithm though then it may make sense to subclass from an existing concrete cell locator class.<br>
<br>HTH,<br>Andy<br><br><div class="gmail_quote">On Mon, Jul 11, 2011 at 11:54 PM, Tharindu De Silva <span dir="ltr"><<a href="mailto:tsameera1@gmail.com">tsameera1@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
Hi Andy,<div><br></div><div> When I go through <i>vtkCellLocator</i> and <i>vtkModifiedBSPTree</i> classes, I find that in addition to finding the containing-cell of a given point (findCell method), they have some extra functionality (e.g. Finding cells which intersect a given line, closest cell within a radius). With what I understood from the paper, it describes an efficient method to locate the cell given a point. I am not sure at this point, whether that algorithm can be extended to improve other functionality such as line intersection. At this point, I can subclass <i>vtkCellLocator</i> and implement FindCell method in that subclass according to the algorithm given in the paper. This is in contrast to the initial idea I had to have another class like <i>vtkCellLocator </i>to implement the given algorithm. Please let me know, what you think about subclassing vtkCellLocator and implementing the FindCell method.</div>
<div><br></div><div>Thank you,</div><div><br></div><div>Regards,</div><div><br></div><div><font color="#888888">Tharindu</font><div><div></div><div class="h5"><br><br><div class="gmail_quote">On Fri, Jul 8, 2011 at 9:18 AM, Andy Bauer <span dir="ltr"><<a href="mailto:andy.bauer@kitware.com" target="_blank">andy.bauer@kitware.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">There's a list of examples at <a href="http://www.vtk.org/Wiki/VTK/Examples" target="_blank">http://www.vtk.org/Wiki/VTK/Examples</a><br>
<br>Doing "ctest -R CellLocator -N" gives me one cell locator test. <br><br>I'm not sure if any of them are specific to unstructured grids but since any VTK grid can be converted to an unstructured grid you should be able to modify them to work as desired.<br>
<font color="#888888">
<br>Andy</font><div><div></div><div><br><br><div class="gmail_quote">On Thu, Jul 7, 2011 at 4:17 PM, Tharindu De Silva <span dir="ltr"><<a href="mailto:tsameera1@gmail.com" target="_blank">tsameera1@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
Hi Andy,<div><br></div><div> Could you please point out to me if there is any existing example that tests cell location in unstructured grids. If there is, I can use that as an initial check during my implementation.</div>
<div><br></div><div>Thank you very much,</div><div><br></div><div><font color="#888888">Tharindu<br><br></font><div class="gmail_quote"><div>On Thu, Jul 7, 2011 at 11:00 AM, Andy Bauer <span dir="ltr"><<a href="mailto:andy.bauer@kitware.com" target="_blank">andy.bauer@kitware.com</a>></span> wrote:<br>
</div><div><div></div><div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">Hi Tharindu,<br><br>I'd be interested in taking a look at what you're doing. Is <a href="http://www.idav.ucdavis.edu/%7Egarth/pdfs/vis10a.pdf" target="_blank">www.idav.ucdavis.edu/~garth/pdfs/vis10a.pdf</a> the paper you're talking about? I haven't gone through it yet so I can't give much feedback now. I'll try to answer any questions you have and if you want me to review your code, feel free to push it to gerrit and add me as a reviewer.<br>
<font color="#888888">
<br>Andy</font><div><div></div><div><br><br><div class="gmail_quote">On Thu, Jul 7, 2011 at 10:50 AM, Berk Geveci <span dir="ltr"><<a href="mailto:berk.geveci@kitware.com" target="_blank">berk.geveci@kitware.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
That sounds like a good plan. I included Will and Andy. They probably<br>
know most about locators and can provide some additional guidance.<br>
What do you think guys?<br>
<br>
PS: Tharindu is the Google Summer of Code student working on<br>
implementing 2 papers from last VisWeek in VTK.<br>
<br>
On Fri, Jul 1, 2011 at 3:12 PM, Jeff Baumes <<a href="mailto:jeff.baumes@kitware.com" target="_blank">jeff.baumes@kitware.com</a>> wrote:<br>
> Berk may have some guidance on the cell locator, or may direct you someone<br>
> else who can help.<br>
> Jeff<br>
><br>
> On Mon, Jun 27, 2011 at 4:17 PM, Tharindu De Silva <<a href="mailto:tsameera1@gmail.com" target="_blank">tsameera1@gmail.com</a>><br>
> wrote:<br>
>><br>
>> HI Jeff and Marcus,<br>
>> I started with the second paper, "Fast memory efficient cell location<br>
>> in unstructered grids for visualization" paper. I have been going through<br>
>> vtkCellLcoator and vtkModifiedBSPTree classes. My plan is to implement<br>
>> another class like this subclassing vtkAbstractCellLocator class. I am<br>
>> sorry, that I was a little slow in understanding the paper due to my lack of<br>
>> background in especially the data structures being used. Now, I understand<br>
>> most of the paper, but there is still more to be done. In the next couple<br>
>> of weeks I will implement that paper.<br>
>> Meanwhile, I requested the author of tick label paper for the data that<br>
>> he used to try an replicate all the results he got. But he didn't get back<br>
>> to me yet. Some figures have a lot of tiny data points that it is hard to<br>
>> trace it visually. I will write an interface that I can trace that data by<br>
>> clicking points and replicate all the results in the paper.<br>
>> Thank you,<br>
>> Tharindu<br>
><br>
</blockquote></div><br>
</div></div></blockquote></div></div></div><br></div>
</blockquote></div><br>
</div></div></blockquote></div><br></div></div></div>
</blockquote></div><br>