Menu
▾
▴
rtftr-devel — Discussions for rtftr developers
You can subscribe to this list here.
| 2008 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(9) |
Oct
(26) |
Nov
(14) |
Dec
(79) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2009 |
Jan
(6) |
Feb
(7) |
Mar
(55) |
Apr
(13) |
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2010 |
Jan
|
Feb
|
Mar
(2) |
Apr
(2) |
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2011 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(2) |
Sep
|
Oct
|
Nov
|
Dec
|
| 2012 |
Jan
|
Feb
|
Mar
(7) |
Apr
(2) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Anjan <roc...@gm...> - 2009-03-17 14:08:28
|
% apporx training time of detector upto 11 stages (140 classifiers selected by AdaBoost) 140 hours %Detection performance data %Image_name actual_face_count detected_face_count false_detection_count detection_time(secs) image_size 0_3.jpg 0 0 0 0.435351 256x192 0_5.jpg 0 1 1 0.469643 256x192 0_11.jpg 0 1 1 0.451269 256x192 0_13.jpg 0 1 1 0.452833 256x192 1_1.jpg 1 1 1 0.429692 256x192 1_4.jpg 1 6 6 0.435552 256x192 1_6.jpg 1 3 2 0.445884 256x192 1_7.jpg 1 2 1 0.441235 256x192 1_8.jpg 1 2 1 0.449807 256x192 1_12.jpg 1 1 0 0.443846 256x192 2_2.jpg 2 3 3 0.445083 256x192 2_9.jpg 2 6 4 0.430605 256x192 2_10.jpg 2 3 1 0.430217 256x192 2_14.jpg 2 4 3 0.439924 256x192 |
|
From: lakesh <lak...@gm...> - 2009-03-17 11:56:51
|
# ------------------------- Performance Log : RTFTR : START -----------------
# Task Id : Recognition Time
# ---------------------------------------------------------------------------
% S.No | Probe Image Filename | Reported Face-Id | Actual Face-Id | Recognition Time (seconds)
1 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v001.jpg 1 1 0.00023043
2 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v002.jpg 1 1 0.000123615
3 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v003.jpg 1 1 0.0001379
4 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v004.jpg 1 1 0.00012082
5 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v005.jpg 1 1 0.000121375
6 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v006.jpg 1 1 0.000120815
7 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v007.jpg 1 1 0.000135655
8 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0001v008.jpg 1 1 0.00012108
9 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v001.jpg 2 2 0.00012058
10 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v002.jpg 2 2 0.00012064
11 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v003.jpg 2 2 0.000120585
12 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v004.jpg 2 2 0.000120545
13 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v005.jpg 2 2 0.00012041
14 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v006.jpg 2 2 0.000120355
15 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v007.jpg 2 2 0.00012128
16 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0002v008.jpg 2 2 0.0001208
17 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v001.jpg 3 3 0.000120895
18 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v002.jpg 3 3 0.000120555
19 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v003.jpg 3 3 0.000120985
20 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v004.jpg 3 3 0.000120925
21 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v005.jpg 3 3 0.000121
22 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v006.jpg 3 3 0.00013277
23 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v007.jpg 3 3 0.00012144
24 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0003v008.jpg 3 3 0.000121155
25 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v001.jpg 4 4 0.000121235
26 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v002.jpg 4 4 0.00012085
27 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v003.jpg 4 4 0.000121065
28 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v004.jpg 4 4 0.0001211
29 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v005.jpg 4 4 0.00012247
30 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v006.jpg 4 4 0.00115923
31 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v007.jpg 4 4 0.00012319
32 /temp/profiler/profileImages/subspacelda/ProjectReportDataSet/probefaces32x32/t0004v008.jpg 4 4 0.000166665
# ------------------------- Performance Log : RTFTR : END |
|
From: Anjan <roc...@gm...> - 2009-03-17 10:56:47
|
Yeah, we'll start extraction from the frame tomorrow and integrate with the recognition module. The result should be ok. Also, the detection time is 0.5 secs, processing the video at 1 FPS won't be a problem. On Tue, Mar 17, 2009 at 4:37 PM, Bibek Shrestha <bib...@gm...>wrote: > The result is really nice. > One thing i notice on the detected images are they are uniform. In case of > frontal faces, you can see that the face detected starts from a little above > the eyes to 2-3pixels below chin. > > I think the results are promising enough, may be we could now train the > recognition module with extraction module? > > Regards > > 2009/3/17 Anjan <roc...@gm...> > >> The attached image shows the detection performance of adaboost detector on >> webcam images. The result is of 11 stages of the cascade and the average >> detection time is around 0.5 seconds on a 256x192 image. >> >> >> -- >> Anjan Nepal >> >> GPO Box 11458 >> Registered Linux User #395963 >> >> >> >> ------------------------------------------------------------------------------ >> Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are >> powering Web 2.0 with engaging, cross-platform capabilities. Quickly and >> easily build your RIAs with Flex Builder, the Eclipse(TM)based development >> software that enables intelligent coding and step-through debugging. >> Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com >> _______________________________________________ >> Rtftr-devel mailing list >> Rtf...@li... >> https://lists.sourceforge.net/lists/listinfo/rtftr-devel >> >> > > > -- > Bibek Shrestha > bibekshrestha at gmail dot com > -- Anjan Nepal GPO Box 11458 Registered Linux User #395963 |
|
From: Bibek S. <bib...@gm...> - 2009-03-17 10:52:56
|
The result is really nice. One thing i notice on the detected images are they are uniform. In case of frontal faces, you can see that the face detected starts from a little above the eyes to 2-3pixels below chin. I think the results are promising enough, may be we could now train the recognition module with extraction module? Regards 2009/3/17 Anjan <roc...@gm...> > The attached image shows the detection performance of adaboost detector on > webcam images. The result is of 11 stages of the cascade and the average > detection time is around 0.5 seconds on a 256x192 image. > > > -- > Anjan Nepal > > GPO Box 11458 > Registered Linux User #395963 > > > > ------------------------------------------------------------------------------ > Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are > powering Web 2.0 with engaging, cross-platform capabilities. Quickly and > easily build your RIAs with Flex Builder, the Eclipse(TM)based development > software that enables intelligent coding and step-through debugging. > Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Bibek Shrestha bibekshrestha at gmail dot com |
|
From: lakesh <lak...@gm...> - 2009-03-16 10:29:17
|
-- Have a nice day |
|
From: Abhishek D. <the...@gm...> - 2009-03-16 06:17:09
|
just fixed all mistakes by talking over telephone :) On Mon, Mar 16, 2009 at 9:26 AM, lakesh <lak...@gm...> wrote: > Hi Abhishek, > > There are so many minor mistakes in the subspace lda portion. I think we > should meet to discuss about the mistakes because there are so many of them > that it is not feasible to write all of them in this email. What do you > say?? > > -- > Have a nice day > > > ------------------------------------------------------------------------------ > Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are > powering Web 2.0 with engaging, cross-platform capabilities. Quickly and > easily build your RIAs with Flex Builder, the Eclipse(TM)based development > software that enables intelligent coding and step-through debugging. > Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Abhishek http://adutta.np.googlepages.com |
|
From: lakesh <lak...@gm...> - 2009-03-16 03:41:48
|
Hi Abhishek, There are so many minor mistakes in the subspace lda portion. I think we should meet to discuss about the mistakes because there are so many of them that it is not feasible to write all of them in this email. What do you say?? -- Have a nice day |
|
From: Abhishek D. <the...@gm...> - 2009-03-13 14:11:58
|
got it On Fri, Mar 13, 2009 at 6:58 PM, lakesh <lak...@gm...> wrote: > Hi Abhishek, > I am sending you my draft report and some data for the plots (using the > database I sent you earlier). I will send you the final report by tomorrow. > About the plots I have the following > > 1) Subspace LDA Matches(Correct and False) > 2) Training Time for two resolutions (64x64 and 128x128) > 3) Recognition time for each probe image for the two resolutions > > I think we should use the average recognition time for each resolution to > compare, instead of recognition time for each probe image. > Also I will try to generate further plots by tommorrow and send you. > > -- > Have a nice day > > > ------------------------------------------------------------------------------ > Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are > powering Web 2.0 with engaging, cross-platform capabilities. Quickly and > easily build your RIAs with Flex Builder, the Eclipse(TM)based development > software that enables intelligent coding and step-through debugging. > Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Abhishek http://adutta.np.googlepages.com |
|
From: lakesh <lak...@gm...> - 2009-03-13 13:13:47
|
# ------------------------- Performance Log : RTFTR : START ----------------- # Task Id : Expected id Vs Obtained id # --------------------------------------------------------------------------- #S.No | Probe Image Name | Recognition Time | Expected Id | Obtained Id 1 /temp/subspacelda/testfaces/1.jpg 3 3 2 /temp/subspacelda/testfaces/2.jpg 3 3 3 /temp/subspacelda/testfaces/3.jpg 3 3 4 /temp/subspacelda/testfaces/4.jpg3 3 3 5 /temp/subspacelda/testfaces/5.jpg 3 3 6 /temp/subspacelda/testfaces/6.jpg 1 1 7 /temp/subspacelda/testfaces/7.jpg 1 1 8 /temp/subspacelda/testfaces/8.jpg 1 1 9 /temp/subspacelda/testfaces/9.jpg 2 2 10 /temp/subspacelda/testfaces/10.jpg 1 1 11 /temp/subspacelda/testfaces/11.jpg 2 2 12 /temp/subspacelda/testfaces/12.jpg 2 3 13 /temp/subspacelda/testfaces/13.jpg 1 1 # ------------------------- Performance Log : RTFTR : END |
|
From: Anjan <roc...@gm...> - 2009-03-13 04:04:02
|
OpenMP basic usage guide page added at wiki https://apps.sourceforge.net/mediawiki/rtftr/index.php?title=OpenMP:Basic_Usage_Guide On Wed, Mar 11, 2009 at 6:13 PM, Abhishek Dutta <the...@gm...>wrote: > Hi Anjan, > Can you create a wiki page that gives basic introduction to OpenMP use for > using multiple cores? You can post the OpenMP based code snippet from > adaboost implementation and add information about compiling such programs. > Don't waste much time doing this. Just create a very short howto for us. > > Email us after you have prepared the wiki page. You can place the wiki page > by linking it with the Main wiki page > > All: > As we will be running all four modules in a single laptop (with 2 cores), > we can use OpenMP (very simple to use - anjan made his code faster by > putting just two #pragma). We can also plot the difference in Face > extraction/recognition time with/without use of OpenMP. Try using OpenMP if > you have extra time and if you wish to optimize your module. > > bye > > -- > Abhishek > http://adutta.np.googlepages.com > > > ------------------------------------------------------------------------------ > Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are > powering Web 2.0 with engaging, cross-platform capabilities. Quickly and > easily build your RIAs with Flex Builder, the Eclipse(TM)based development > software that enables intelligent coding and step-through debugging. > Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Anjan Nepal GPO Box 11458 Registered Linux User #395963 |
|
From: Abhishek D. <the...@gm...> - 2009-03-11 14:07:48
|
Hi all, I have added some sample plots to this wiki page for FE modules http://apps.sourceforge.net/mediawiki/rtftr/index.php?title=Guidelines_for_preparing_final_year_project_report Take a careful look before generating statistical data These plots will give you an idea about the nature of raw data you need to send me on March 15, 2009 to generate plots for the report. I will add sample plots for FE modules tomorrow. bye -- Abhishek http://adutta.np.googlepages.com |
|
From: Abhishek D. <the...@gm...> - 2009-03-11 12:28:58
|
Hi Anjan, Can you create a wiki page that gives basic introduction to OpenMP use for using multiple cores? You can post the OpenMP based code snippet from adaboost implementation and add information about compiling such programs. Don't waste much time doing this. Just create a very short howto for us. Email us after you have prepared the wiki page. You can place the wiki page by linking it with the Main wiki page All: As we will be running all four modules in a single laptop (with 2 cores), we can use OpenMP (very simple to use - anjan made his code faster by putting just two #pragma). We can also plot the difference in Face extraction/recognition time with/without use of OpenMP. Try using OpenMP if you have extra time and if you wish to optimize your module. bye -- Abhishek http://adutta.np.googlepages.com |
|
From: Abhishek D. <the...@gm...> - 2009-03-06 01:53:11
|
Hi Lakesh, We need a basic GUI to show the faces being tracked and their respective identity. We will discuss about this today in college and make plans for developing a basic GUI. bye On Fri, Mar 6, 2009 at 7:16 AM, lakesh <lak...@gm...> wrote: > Hi all, > > I was just thinking that we should also have a basic GUI for the final > presentation. So what do you guys think about this?? Shall we start working > on it after we get the integration done? > > -- > Have a nice day > > > ------------------------------------------------------------------------------ > Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, > CA > -OSBC tackles the biggest issue in open source: Open Sourcing the > Enterprise > -Strategies to boost innovation and cut costs with open source > participation > -Receive a $600 discount off the registration fee with the source code: > SFAD > http://p.sf.net/sfu/XcvMzF8H > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Abhishek http://adutta.np.googlepages.com |
|
From: lakesh <lak...@gm...> - 2009-03-06 01:47:52
|
Hi all, I was just thinking that we should also have a basic GUI for the final presentation. So what do you guys think about this?? Shall we start working on it after we get the integration done? -- Have a nice day |
|
From: Abhishek D. <the...@gm...> - 2009-03-06 00:52:21
|
That's great We still have around 8 days to finish the training process. bye On Thu, Mar 5, 2009 at 10:59 PM, Anjan <roc...@gm...> wrote: > Hi all, > Adaboost training has been started from Thu, 5 Mar, 2009 at 22:30. We can > regularly view the status, and accordingly make changes to try to complete > the training according to our final project presentation schedule. > > initial training images > * faces 750 x 2 (x2 = mirrored) > * nonfaces 1500 x 2 > for later stages > * faces (constant as above) > * nonfaces 2000 x 2 > > To speed up the training, we can reduce the number of nonfaces that will be > created, but this might affect the accuracy of the application. > > We can regularly view the status of the training from ssh. (thanks to > bibek's work :) ) > > bye > > > -- > Anjan Nepal > > GPO Box 11458 > Registered Linux User #395963 > > -- Abhishek http://adutta.np.googlepages.com |
|
From: Anjan <roc...@gm...> - 2009-03-05 17:14:19
|
Hi all, Adaboost training has been started from Thu, 5 Mar, 2009 at 22:30. We can regularly view the status, and accordingly make changes to try to complete the training according to our final project presentation schedule. initial training images * faces 750 x 2 (x2 = mirrored) * nonfaces 1500 x 2 for later stages * faces (constant as above) * nonfaces 2000 x 2 To speed up the training, we can reduce the number of nonfaces that will be created, but this might affect the accuracy of the application. We can regularly view the status of the training from ssh. (thanks to bibek's work :) ) bye -- Anjan Nepal GPO Box 11458 Registered Linux User #395963 |
|
From: Anjan <roc...@gm...> - 2009-03-01 19:35:17
|
Hi Bibek, The red circles include the false positives that contain plain surfaces. I had once read in a paper that this can be reduced by calculating variance of the window and rejecting the ones which have variance less than some threshold. And, in the given image, i think the detection is the rectange which includes the eyes and the mouth. So, the detection has small portion of mouth and eye cut. What do u think? 2009/3/1 Bibek Shrestha <bib...@gm...> > Check the red circles > > -- > Bibek Shrestha > bibekshrestha at gmail dot com > > > ------------------------------------------------------------------------------ > Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, > CA > -OSBC tackles the biggest issue in open source: Open Sourcing the > Enterprise > -Strategies to boost innovation and cut costs with open source > participation > -Receive a $600 discount off the registration fee with the source code: > SFAD > http://p.sf.net/sfu/XcvMzF8H > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Anjan Nepal GPO Box 11458 Registered Linux User #395963 |
|
From: Abhishek D. <the...@gm...> - 2009-03-01 14:56:51
|
Hi all, We have a meeting with our supervisors on March 04, 2009 at 3:00PM in our Department. Reminder: We all are meeting tomorrow in college at 10:00AM bye -- Abhishek http://adutta.np.googlepages.com |
|
From: Abhishek D. <the...@gm...> - 2009-03-01 14:34:31
|
Hi Bibek, I am not aware about any rejection techniques. However, if you pass those windows (containing wall) to lda or kepenekci face recognition module, then we can easily tell you if it has features of a human face or not. LDA and Kepenekci modules perform the recognition process at very high speed (its speed will increase even further is we use OpenMP to parallelize stuffs). So for now, you can leave it to FR modules to reject non-faces (for now only). bye On Sun, Mar 1, 2009 at 3:53 PM, Bibek Shrestha <bib...@gm...>wrote: > Neural networks detection is working properly right now. > One problem that has arized is.. > > Since we are doing histogram equalization in the image, this causes a lot > of noise to smoother image. > > For example say an image subwindow contains very small variations around > range of 240 to 250 only (consider a wall), such image after histogram > equilization will have a lot of noise after histogram equilization and > produce face like structure. > > i have been searching for possible rejection techniques to reject these > windows beforehand, by checking the pixel variation but still haven't found > a permanent solution > > Anjan, you once suggested me about such similar technique, do you have any > idea about it? > > If any one knows any technique, it would greatly speed up the filtering > process. > > Cheers > > -- > Bibek Shrestha > bibekshrestha at gmail dot com > > > ------------------------------------------------------------------------------ > Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, > CA > -OSBC tackles the biggest issue in open source: Open Sourcing the > Enterprise > -Strategies to boost innovation and cut costs with open source > participation > -Receive a $600 discount off the registration fee with the source code: > SFAD > http://p.sf.net/sfu/XcvMzF8H > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Abhishek http://adutta.np.googlepages.com |
|
From: Dr. J. T. <jyo...@gm...> - 2009-03-01 11:12:50
|
Hi, I hope you all did well in the exam, and are back to work after rest of "a day or two". It seems you have met only once after exam - so you may already be planning to meet this week too. If you are meeting on Wednesday at the department, I and Sharad both will be available in the afternoon. It is good to see the development going smoothly. It is now time to do the documentation and prepare presentations too. You may be having the mid-term (?) progress/defense soon, so I would like to review your presentation for the same. Jyoti From: the...@gm... [mailto:the...@gm...] On Behalf Of Abhishek Dutta Sent: Wednesday, November 26, 2008 9:00 PM To: Dr. Jyoti Tandukar Cc: RTFTR-devel Mailing List Subject: Real Time Face Tracking & Recognition (RTFTR) project details Dear Sir, The details of the progress with "rtftr" project can be monitored using the following sites: 1. http://rtftr.sourceforge.net - the project website (for users curious about our project) 2. http://apps.sourceforge.net/mediawiki/rtftr/index.php?title=Main_Page - the rtftr team collaborates on project schedules, documents, etc using this WIKI 3. http://sourceforge.net/mailarchive/forum.php?forum_name=rtftr-devel - rtftr developers communicate via this mailing list 4. http://rtftr.svn.sourceforge.net/viewvc/rtftr/trunk - subversion repository for rtftr development code . We all will brief you about the progress of "rtftr" project on Sunday at 1:35 PM. Thank you RTFTR team -- Abhishek http://adutta.np.googlepages.com |
|
From: Bibek S. <bib...@gm...> - 2009-03-01 10:08:53
|
Neural networks detection is working properly right now. One problem that has arized is.. Since we are doing histogram equalization in the image, this causes a lot of noise to smoother image. For example say an image subwindow contains very small variations around range of 240 to 250 only (consider a wall), such image after histogram equilization will have a lot of noise after histogram equilization and produce face like structure. i have been searching for possible rejection techniques to reject these windows beforehand, by checking the pixel variation but still haven't found a permanent solution Anjan, you once suggested me about such similar technique, do you have any idea about it? If any one knows any technique, it would greatly speed up the filtering process. Cheers -- Bibek Shrestha bibekshrestha at gmail dot com |
|
From: Abhishek D. <the...@gm...> - 2009-02-24 17:21:47
|
Thanx lakesh You saved everybody's time :) bye On Tue, Feb 24, 2009 at 10:43 PM, lakesh <lak...@gm...> wrote: > Hi guys, > Use this link for creating a shared object file in kdevelop > > > http://www.kdevelop.org/mediawiki/index.php/FAQ#How_to_create_a_.so_library.3F > > Among the two options given in the link I used the second one (libtool > based method). It creates a .la file, then you need to install the library > in the form of .so . Follow the procedure as given in the link. Then it will > create a .la file. Then u need to install it in the form of .so file in the > /usr/local/lib folder. For that just right click on the subproject and the > install option. It will create the required so file in the /usr/local/lib > path. One more thing u need to have root permission to create any file in > the /usr/local/lib path. Though there is a install as root option I had a > /bin/sh:kdesu not found error. So just run kdevelop as root and then click > on install and the shared library will be created. I hope this would be > helpful to you guys so that you don't need to waste time in googling. > -- > Have a nice day > > > ------------------------------------------------------------------------------ > Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, > CA > -OSBC tackles the biggest issue in open source: Open Sourcing the > Enterprise > -Strategies to boost innovation and cut costs with open source > participation > -Receive a $600 discount off the registration fee with the source code: > SFAD > http://p.sf.net/sfu/XcvMzF8H > _______________________________________________ > Rtftr-devel mailing list > Rtf...@li... > https://lists.sourceforge.net/lists/listinfo/rtftr-devel > > -- Abhishek http://adutta.np.googlepages.com |
|
From: lakesh <lak...@gm...> - 2009-02-24 16:58:43
|
Hi guys, Use this link for creating a shared object file in kdevelop http://www.kdevelop.org/mediawiki/index.php/FAQ#How_to_create_a_.so_library.3F Among the two options given in the link I used the second one (libtool based method). It creates a .la file, then you need to install the library in the form of .so . Follow the procedure as given in the link. Then it will create a .la file. Then u need to install it in the form of .so file in the /usr/local/lib folder. For that just right click on the subproject and the install option. It will create the required so file in the /usr/local/lib path. One more thing u need to have root permission to create any file in the /usr/local/lib path. Though there is a install as root option I had a /bin/sh:kdesu not found error. So just run kdevelop as root and then click on install and the shared library will be created. I hope this would be helpful to you guys so that you don't need to waste time in googling. -- Have a nice day |
|
From: Abhishek D. <the...@gm...> - 2009-02-23 17:19:33
|
Hi all, check the tickets at http://apps.sourceforge.net/trac/rtftr/report/1 Update the ticket status regularly to reflect your progress. bye -- Abhishek http://adutta.np.googlepages.com |
|
From: Abhishek D. <the...@gm...> - 2009-02-23 16:15:43
|
Hi all, The workflow (as decided in today's meeting) has been posted in trac's wiki at http://apps.sourceforge.net/trac/rtftr/ We will be using trac for only issue tracking and media wiki (the old wiki that we have been using) will be used for wiki. Trac's wiki is good but lots of documents related to RTFTR exists in old wiki. So we will _NOT_ switch to trac's wiki. The links of various resources have been update in our project's website at: http://rtftr.sourceforge.net We will be using these version numbering to indicate different stages of rtftr's development: - *RTFTR version 1.0 will be submitted to Dept. of Electronics & Computer Engineering <http://doece.ioe.edu.np/>, Pulchowk Campus<http://www.ioe.edu.np/>as final year project * - *RTFTR version 2.0 will be submitted to Open Software Challenge (OSC) - 2009 <http://www.d2labs.org/oscn/index.html>* The tickets (assigning tasks to each one of us) will be posted in trac very soon. Bye -- Abhishek http://adutta.np.googlepages.com |
23 messages has been excluded from this view by a project administrator.
