<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Oct 15, 2015, at 12:52 AM, Michael Rosen <<a href="mailto:michael.rosen@gmail.com" class="">michael.rosen@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class="">Umm, yes, there's no "Grid" in the config (all commented out), but I misspoke about it failing somewhere else. It fails in the same place. Looks like the OCV plugin is asking CV for a "SURF" FeatureDetector, the implementation in CV is the "different" error I mentioned. Sorry for the confusion.</div></div></blockquote><div><br class=""></div><div>Thanks for the clarification. I’ve never run into that issue when not using Grid. One other possibility is that your OpenCV was built without the “Non-Free” components turned on. SURF is part of “non-free” because of patent issues.</div><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class=""><br class=""></div><div class="">Sounds like to get a Windows Port we need to either fix the obscure bug in CV2 or migrate MapTK to CV3.<br class=""></div></div></div></blockquote><div><br class=""></div><div>Yes, but fixing the OpenCV 2.4.X bug and getting is accepted upstream is unlikely because the OpenCV team didn’t fix it for 2 years and then decided to remove the feature in 3.0. We could actually work around this issue in a way that would work with both OpenCV 2.4.X and 3.0, but it involves a bit of reworking our current code and hard coding a list of OpenCV algorithms that MAP-Tk knows about. It’s not as elegant as our current approach, but it might be the direction we need to take.</div><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class=""><div class=""><br class=""></div><div class="">I downloaded the VM you mentioned and was able to walk through the tutorial ... thanks for taking the effort to do that. </div><div class=""><br class=""></div><div class="">Is there a way to use the existing MapTK tools or SDK to create a rectified image from the overlapping collection?</div></div></div></div></blockquote><div><br class=""></div><div>Using only MAP-Tk code there is not currently code to warp all of the images to a common image plane and create a mosaic (I’m assuming that’s what you are asking about). However there is code in another KWIVER project that I’ve been using for this:</div><div><br class=""></div><div><a href="https://github.com/Kitware/vibrant/blob/master/tools/warp_video.cxx" class="">https://github.com/Kitware/vibrant/blob/master/tools/warp_video.cxx</a></div><div><br class=""></div><div>It uses the homographies produced by the MAP-Tk feature tracking tool to warp all of the images. The ultimate plan is to move a version of this tool into MAP-Tk. In fact, I could probably write a very simple version of this tool as a Python script using OpenCV, I just haven’t gotten to it yet.</div><div><br class=""></div><div>—Matt</div><div><br class=""></div><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class=""><div class=""><br class=""></div><div class="">msr</div><div class=""><br class=""></div><div class=""><br class=""></div></div></div><div class="gmail_extra"><br class=""><div class="gmail_quote">On Wed, Oct 14, 2015 at 7:49 PM, Matthew Leotta <span dir="ltr" class=""><<a href="mailto:matt.leotta@kitware.com" target="_blank" class="">matt.leotta@kitware.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word" class="">Michael,<div class=""><br class=""></div><div class="">Can you confirm that “Grid” no longer appears anywhere in your config? I’m not sure how this code path would be reached if you are no longer configured to use the GridAdaptedFeatureDetector.<br class=""><div class=""><br class=""></div><div class="">The most relevant OpenCV ticket is this one:</div><div class=""><br class=""></div><div class=""><a href="http://code.opencv.org/issues/2700" target="_blank" class="">http://code.opencv.org/issues/2700</a></div><div class=""><br class=""></div><div class="">You can see the solution is “ok, we removed this feature from 3.0”, so they no longer need to address the problem. Another related ticket is here:</div><div class=""><br class=""></div><div class=""><a href="http://code.opencv.org/issues/4129" target="_blank" class="">http://code.opencv.org/issues/4129</a></div><div class=""><br class=""></div><div class="">This one suggests that going forward (starting in OpenCV 3.0) one needs to explicitly call create functions on the class instance, instead of creating by name. This means that when we update MAP-Tk to use OpenCV 3.0 we will also need reproduce this lookup by name on the MAP-Tk side.</div><div class=""><br class=""></div><div class="">Also, it seems that the GridAdaptiveFeatureDetector itself (and all the other adaptive feature detector modifiers) have gone missing in OpenCV 3.0 as noted here:</div><div class=""><br class=""></div><div class=""><a href="http://answers.opencv.org/question/42490/gridadaptedfeaturedetector-missing-in-opencv-30/" target="_blank" class="">http://answers.opencv.org/question/42490/gridadaptedfeaturedetector-missing-in-opencv-30/</a></div><div class=""><br class=""></div><div class="">I started a branch to upgrade to OpenCV 3.0, but got stuck with these issues. In summary, there doesn’t appear to be a simple fix here. I’d be happy to have someone volunteer to tackle this issue.</div><div class=""><br class=""></div><div class="">—Matt</div><div class=""><div class="h5"><div class=""><br class=""></div><div class=""><br class=""></div><div class=""><br class=""><div class=""><blockquote type="cite" class=""><div class="">On Oct 14, 2015, at 6:42 PM, Michael Rosen <<a href="mailto:michael.rosen@gmail.com" target="_blank" class="">michael.rosen@gmail.com</a>> wrote:</div><br class=""><div class=""><div dir="ltr" class="">Ah. I missed the change. Did that ... and it did help but then it does the same thing somewhere else:<div class=""><br class=""></div><div class=""><div class=""> <span style="white-space:pre-wrap" class=""> </span>msvcr120.dll!memchr(unsigned char * buf, unsigned char chr, unsigned long cnt) Line 101<span style="white-space:pre-wrap" class=""> </span>Unknown</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>opencv_features2d249.dll!std::basic_string<char,std::char_traits<char>,std::allocator<char> >::find(const char * _Ptr, unsigned int _Off, unsigned int _Count) Line 1908<span style="white-space:pre-wrap" class=""> </span>C++</div><div class="">><span style="white-space:pre-wrap" class=""> </span>opencv_features2d249.dll!cv::FeatureDetector::create(const std::basic_string<char,std::char_traits<char>,std::allocator<char> > & detectorType) Line 93<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::detect_features::priv::default_detector() Line 76<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::detect_features::priv::priv() Line 63<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::detect_features::detect_features() Line 94<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::algo::algorithm_impl<maptk::ocv::detect_features,maptk::algo::detect_features>::register_self(maptk::registrar & reg) Line 370<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::register_algorithms(maptk::registrar & reg) Line 70<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv_plugin.dll!register_algo_impls(maptk::registrar & reg) Line 44<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv_plugin.dll!private_register_algo_impls(maptk::registrar & reg) Line 59<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::impl::register_from_module(boost::filesystem::path module_path) Line 293<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::impl::load_modules_in_directory(boost::filesystem::path dir_path, std::basic_string<char,std::char_traits<char>,std::allocator<char> > name) Line 198<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::impl::load_from_search_paths(std::basic_string<char,std::char_traits<char>,std::allocator<char> > name) Line 146<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::register_plugins(std::basic_string<char,std::char_traits<char>,std::allocator<char> > name) Line 409<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!maptk_main(int argc, const char * * argv) Line 236<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!main(int argc, const char * * argv) Line 489<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!__tmainCRTStartup() Line 626<span style="white-space:pre-wrap" class=""> </span>C</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!mainCRTStartup() Line 466<span style="white-space:pre-wrap" class=""> </span>C</div></div><div class=""><br class=""></div><div class=""><div class="">Ptr<FeatureDetector> FeatureDetector::create( const string& detectorType )</div><div class="">{</div><div class=""> if( detectorType.find("Grid") == 0 ) /// AV here.</div><div class=""> {</div><div class=""> return new GridAdaptedFeatureDetector(FeatureDetector::create(</div><div class=""> detectorType.substr(strlen("Grid"))));</div><div class=""> }</div></div><div class=""><br class=""></div><div class=""><br class=""></div><div class="">Thanks very much for the Linux pointers. I'll try those.</div><div class=""><br class=""></div><div class="">At some point, I'll need to speak to the prospects for making this work on Windows. Can you point me to a ticket for the "<span style="font-size:12.8px" class="">OpenCV’s algorithm factory code on Windows" issue you mentioned or otherwise comment on the prospects for a fix?</span></div><div class=""><span style="font-size:12.8px" class=""><br class=""></span></div><div class=""><span style="font-size:12.8px" class="">msr</span></div></div><div class="gmail_extra"><br class=""><div class="gmail_quote">On Wed, Oct 14, 2015 at 7:09 AM, Matthew Leotta <span dir="ltr" class=""><<a href="mailto:matt.leotta@kitware.com" target="_blank" class="">matt.leotta@kitware.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word" class="">It’s not just commenting out lines that is needed. You also have to modify the other lines to remove the Features2D.Grid wrapper around the Features2D.SURF. I’ve illustrated what needs to change in my previous e-mail. That said, if you can use Linux that’s a better solution.<div class=""><br class=""></div><div class="">There are a few options for the Linux route:</div><div class=""><br class=""></div><div class="">1) Keith or someone else can provide guidance on using Kwiver to build boost and other dependencies. I think a lot of that has been cleaned up in the last couple of weeks.</div><div class=""><br class=""></div><div class="">2) You can use the pre-built VM that comes with the CVPR tutorial. That has MAP-Tk pre-built and ready to run. You can either download the VM image or use Vagrant to reproduce it yourself.</div><div class=""><br class=""></div><div class="">3) Depending on your Linux distro, you may be able to install system packages for boost and other dependencies. For example, the VM mentioned above does this using Ubuntu 14.04. If you happen to be using Ubuntu, you can look through the Vagrant setup scripts to see exactly how to install all dependencies and build MAP-Tk.</div><div class=""><br class=""></div><div class="">—Matt<div class=""><div class=""><br class=""><div class=""><br class=""></div><div class=""><br class=""><div class=""><blockquote type="cite" class=""><div class="">On Oct 14, 2015, at 9:58 AM, Michael Rosen <<a href="mailto:michael.rosen@gmail.com" target="_blank" class="">michael.rosen@gmail.com</a>> wrote:</div><br class=""><div class=""><div dir="ltr" class="">I commented out lines from clif_track.conf as follows but it failed as before:<div class=""><div class=""># The OpenCV cv::Algorithm type to use for 'detector'.</div><div class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:detector:type = Feature2D.SURF</div><div class=""><br class=""></div><div class="">#feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:gridCols = 4</div><div class="">#feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:gridRows = 6</div><div class="">#feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:maxTotalKeypoints = 1000</div></div><div class=""><br class=""></div><div class="">I had previously tried building Kwiver on Linux but it failed to build Boost. Can you recommend a way forward for me?</div><div class=""><br class=""></div><div class="">Thanks.</div><div class=""><br class=""></div><div class="">msr</div></div><div class="gmail_extra"><br class=""><div class="gmail_quote">On Wed, Oct 14, 2015 at 6:42 AM, Matthew Leotta <span dir="ltr" class=""><<a href="mailto:matt.leotta@kitware.com" target="_blank" class="">matt.leotta@kitware.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word" class="">Michael,<div class=""><br class=""></div><div class="">This is likely related persistent issues with OpenCV’s algorithm factory code on Windows. For whatever reason, the OpenCV team could never seem to work the kinks out of this on Windows, but it works fine on other platforms. I was really hoping they were going to fix it once and for all in OpenCV 3.0, but instead they decided it was too much hassle and they just removed algorithm factories altogether. So OpenCV no longer supports creating algorithms at run-time from string names. This makes me sad, because we had a nice system in MAP-Tk of allowing OpenCV algorithms to be created and configured on the fly using the MAP-Tk configuration file. If we upgrade MAP-Tk to use OpenCV 3.0 we will have to hard code a fixed set of algorithm choices and configuration parameters.</div><div class=""><br class=""></div><div class="">The problem seems to mostly manifest itself when you use nested OpenCV algorithm on Windows. In this case, the example config file contains:</div><div class=""><br class=""></div><div class="">-----------------------<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:detector:Feature2D.SURF:extended = false<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:detector:Feature2D.SURF:hessianThreshold = 100<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:detector:Feature2D.SURF:nOctaveLayers = 3<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:detector:Feature2D.SURF:nOctaves = 4<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:detector:Feature2D.SURF:upright = true<br class=""><br class=""># The OpenCV cv::Algorithm type to use for 'detector'.<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:detector:type = Feature2D.SURF<br class=""><br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:gridCols = 4<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:gridRows = 6<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.Grid:maxTotalKeypoints = 1000<br class=""><br class=""># The OpenCV cv::Algorithm type to use for 'detector'.<br class="">feature_tracker:core:feature_detector:ocv:detector:type = Feature2D.Grid</div><div class="">----------------------- </div><div class=""><br class=""></div><div class="">This might run if you remove the OpenCV grid adaptive detector layer like this:</div><div class=""><br class=""></div><div class=""><div class="">----------------------- <br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.SURF:extended = false<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.SURF:hessianThreshold = 100<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.SURF:nOctaveLayers = 3<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.SURF:nOctaves = 4<br class="">feature_tracker:core:feature_detector:ocv:detector:Feature2D.SURF:upright = true<br class=""><br class=""># The OpenCV cv::Algorithm type to use for 'detector'.<br class="">feature_tracker:core:feature_detector:ocv:detector:type = Feature2D.SURF</div><div class="">----------------------- </div></div><div class=""><br class=""></div><div class="">I would expect this to run (hopefully), but it is likely that it may not produce ideal results.</div><div class=""><br class=""></div><div class="">—Matt</div><div class=""><div class=""><div class=""><br class=""></div><div class=""><br class=""><div class=""><blockquote type="cite" class=""><div class="">On Oct 14, 2015, at 9:13 AM, Michael Rosen <<a href="mailto:michael.rosen@gmail.com" target="_blank" class="">michael.rosen@gmail.com</a>> wrote:</div><br class=""><div class=""><div dir="ltr" class="">Presentation and test data look great. Sadly, it's not working for me on Windows. maptk_track_features posts an AV. Can you provide some insight into what's wrong?<div class=""><br class=""></div><div class="">Attempting to follow along with the example, I ran this:</div><div class=""><br class=""></div><div class=""><div class=""> C:\dev\cvpr2015-opensfm-master\Exercises\maptk>C:\dev\maptk\bin\maptk_track_features.exe -c clif_track.conf</div><div class=""><br class=""></div><div class="">I think the debug spew below looks ok:</div><div class=""><br class=""></div><div class="">[DEBUG][maptk::algorithm_plugin_manager::register_plugins] Dynamically loading plugin impls</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::load_from_search_paths] Loading plugins in search paths</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::load_modules_in_directory] Loading modules in directory: "C:/dev/maptk/lib/maptk"</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] Starting plug-in interfacing for module file: "C:/dev/maptk/lib</div><div class="">/maptk\maptk_core_plugin.dll"</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] Loaded module: 73EE0000</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] Looking for algorithm impl registration function: private_regis</div><div class="">ter_algo_impls</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] -> returned function address: 73EE1163</div><div class="">[DEBUG][maptk::Implementation Registration] Registering algorithm implementations from module 'maptk_core'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'close_loops:bad_frames_only' for def type 'close_loops'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'bad_frames_only' for def type 'close_loops'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'close_loops:multi_method' for def type 'close_loops'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'multi_method' for def type 'close_loops'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'compute_ref_homography:core' for def type 'compute_ref_homography'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'core' for def type 'compute_ref_homography'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'convert_image:bypass' for def type 'convert_image'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'bypass' for def type 'convert_image'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'filter_features:magnitude' for def type 'filter_features'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'magnitude' for def type 'filter_features'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'bundle_adjust:hierarchical' for def type 'bundle_adjust'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'hierarchical' for def type 'bundle_adjust'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'match_features:homography_guided' for def type 'match_features'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'homography_guided' for def type 'match_features'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'track_features:core' for def type 'track_features'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'core' for def type 'track_features'</div><div class="">[DEBUG][maptk::algorithm_plugin_interface_macros::REGISTRATION_SUMMARY] Registered 8 of 8 algorithms</div><div class=""> (@C:\dev\maptk\maptk\plugins\core\register_algorithms.cxx)</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] -> Successfully called registration func</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] Starting plug-in interfacing for module file: "C:/dev/maptk/lib</div><div class="">/maptk\maptk_ocv_plugin.dll"</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] Loaded module: 73EA0000</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] Looking for algorithm impl registration function: private_regis</div><div class="">ter_algo_impls</div><div class="">[DEBUG][maptk::algorithm_plugin_manager::impl::register_from_module] -> returned function address: 73EA1163</div><div class="">[DEBUG][maptk::Implementation Registration] Registering algorithm implementations from module 'maptk_ocv'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'analyze_tracks:ocv' for def type 'analyze_tracks'</div><div class="">[DEBUG][maptk::registrar] Registering algo implementation 'ocv' for def type 'analyze_tracks'</div></div><div class=""><br class=""></div><div class="">But then right after that, I get an Access Violation with this call stack. What's happening?</div><div class=""><br class=""></div><div class=""><div class=""> <span style="white-space:pre-wrap" class=""> </span>msvcr120.dll!memchr(unsigned char * buf, unsigned char chr, unsigned long cnt) Line 101<span style="white-space:pre-wrap" class=""> </span>Unknown</div><div class="">><span style="white-space:pre-wrap" class=""> </span>opencv_features2d249.dll!std::basic_string<char,std::char_traits<char>,std::allocator<char> >::find(const char * _Ptr, unsigned int _Off, unsigned int _Count) Line 1908<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>opencv_features2d249.dll!cv::FeatureDetector::create(const std::basic_string<char,std::char_traits<char>,std::allocator<char> > & detectorType) Line 93<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::detect_features::priv::default_detector() Line 76<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::detect_features::priv::priv() Line 63<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::detect_features::detect_features() Line 94<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::algo::algorithm_impl<maptk::ocv::detect_features,maptk::algo::detect_features>::register_self(maptk::registrar & reg) Line 370<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv.dll!maptk::ocv::register_algorithms(maptk::registrar & reg) Line 70<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv_plugin.dll!register_algo_impls(maptk::registrar & reg) Line 44<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_ocv_plugin.dll!private_register_algo_impls(maptk::registrar & reg) Line 59<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::impl::register_from_module(boost::filesystem::path module_path) Line 293<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::impl::load_modules_in_directory(boost::filesystem::path dir_path, std::basic_string<char,std::char_traits<char>,std::allocator<char> > name) Line 198<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::impl::load_from_search_paths(std::basic_string<char,std::char_traits<char>,std::allocator<char> > name) Line 146<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_apm.dll!maptk::algorithm_plugin_manager::register_plugins(std::basic_string<char,std::char_traits<char>,std::allocator<char> > name) Line 409<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!maptk_main(int argc, const char * * argv) Line 236<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!main(int argc, const char * * argv) Line 489<span style="white-space:pre-wrap" class=""> </span>C++</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!__tmainCRTStartup() Line 626<span style="white-space:pre-wrap" class=""> </span>C</div><div class=""> <span style="white-space:pre-wrap" class=""> </span>maptk_track_features.exe!mainCRTStartup() Line 466<span style="white-space:pre-wrap" class=""> </span>C</div></div><div class=""><br class=""></div><div class=""><br class=""></div><div class="">The offending code seems to be this:</div><div class=""><div class="">/// create the default feature detector</div><div class=""> static cv::Ptr<cv::FeatureDetector> default_detector()</div><div class=""> {</div><div class=""> cv::Ptr<cv::FeatureDetector> det;</div><div class=""> // try the SURF detector first</div><div class=""> det = cv::FeatureDetector::create("SURF"); //////// Access Violation here.</div><div class=""> if( !det )</div><div class=""> {</div><div class=""> // if SURF is not available (nonfree not built) use ORB</div><div class=""> det = cv::FeatureDetector::create("ORB");</div><div class=""> }</div><div class=""> return det;</div><div class=""> }</div></div></div><div class="gmail_extra"><br class=""><div class="gmail_quote">On Tue, Oct 6, 2015 at 2:47 PM, Michael Rosen <span dir="ltr" class=""><<a href="mailto:michael.rosen@gmail.com" target="_blank" class="">michael.rosen@gmail.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr" class="">Thanks very much. This looks really helpful. I'll give it a spin and let you know how it goes.<div class=""><br class=""></div><div class="">msr</div></div><div class=""><div class=""><div class="gmail_extra"><br class=""><div class="gmail_quote">On Tue, Oct 6, 2015 at 2:01 PM, Matthew Leotta <span dir="ltr" class=""><<a href="mailto:matt.leotta@kitware.com" target="_blank" class="">matt.leotta@kitware.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word" class=""><br class=""><div class=""><span class=""><blockquote type="cite" class=""><div class="">On Oct 6, 2015, at 3:38 PM, Michael Rosen <<a href="mailto:michael.rosen@gmail.com" target="_blank" class="">michael.rosen@gmail.com</a>> wrote:</div><br class=""><div class=""><div dir="ltr" class="">Thanks for this. It compiles and the assert is gone. <div class=""><br class=""></div><div class="">To be clear, here's what I did:</div><blockquote style="margin:0 0 0 40px;border:none;padding:0px" class=""><div class="">git clone <a href="https://github.com/Kitware/maptk.git" target="_blank" class="">https://github.com/Kitware/maptk.git</a></div><div class=""><div class="">cd .\maptk</div></div><div class=""><div class="">git fetch origin pull/88/head:dev/camera-pass-by-reference</div></div><div class=""><div class="">git checkout dev/camera-pass-by-reference</div></div><div class=""><div class="">cmake -G "NMake Makefiles" -DBOOST_ROOT=C:\dev\fletch\install -DEIGEN3_INCLUDE_DIR="C:\Program Files (x86)\Eigen\include\eigen3"</div></div><div class=""><div class="">nmake</div></div><div class=""><div class="">nmake install</div></div></blockquote></div></div></blockquote><div class=""><br class=""></div><div class=""><br class=""></div></span><div class="">Interesting, so you didn’t need to do any of the defines to disable byte alignment for Eigen? Were you building master or release before?</div><div class=""><br class=""></div><div class="">By the way, I have now merged that dev/camera-pass-by-reference branch into both the release and master branches of MAP-Tk, so you should be able to use one of the primary branches again.</div><span class=""><div class=""><br class=""></div><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><blockquote style="margin:0 0 0 40px;border:none;padding:0px" class=""><div class=""><br class=""></div></blockquote><div class=""><div class="">Is there a simple workflow you can pass that will demonstrate use of some of the utilities. For example, suppose I have a collection of overlapping images, what can I do with them?</div></div></div></div></blockquote><div class=""><br class=""></div></span><div class="">I’ve been meaning to write up a tutorial on this, but haven’t gotten to it yet. Part of this is because there are big API changes coming soon as well as improvements in algorithms. There is a bunch of stuff building up that I can’t release yet until I get approval from AFRL. A preview of those changes is on the kwiver-integration branch. We are pulling a lot of core guts of MAP-Tk out into a new repository named VITAL (<a href="https://github.com/kitware/vital" target="_blank" class="">https://github.com/kitware/vital</a>) that is shared across KWIVER projects. I don’t recommend building that on Windows just yet. But Keith Fieldhouse is working on a KWIVER “super build” that will use CMake to make the various projects build together easily.</div><div class=""><br class=""></div><div class="">Anyway, the best I can do at this point is to point you to a tutorial I gave back in June at the CVPR conference:</div><div class=""><br class=""></div><div class=""><a href="http://www.kitware.com/cvpr2015-tutorial.html" target="_blank" class="">http://www.kitware.com/cvpr2015-tutorial.html</a></div><div class=""><br class=""></div><div class="">This used a version of MAP-Tk similar to the current release branch. There are examples for applying MAP-Tk to a sample dataset. A Linux VM image is provided with all the software and data pre-installed, but you can also get the config files here:</div><div class=""><br class=""></div><div class=""><a href="https://github.com/mleotta/cvpr2015-opensfm/tree/master/Exercises/maptk" target="_blank" class="">https://github.com/mleotta/cvpr2015-opensfm/tree/master/Exercises/maptk</a></div><div class=""><br class=""></div><div class="">and the sample data here:</div><div class=""><br class=""></div><div class=""><a href="https://github.com/mleotta/cvpr2015-opensfm/tree/master/Data/CLIF_2007" target="_blank" class="">https://github.com/mleotta/cvpr2015-opensfm/tree/master/Data/CLIF_2007</a></div><div class=""><br class=""></div><div class="">and slides explaining what to do here:</div><div class=""><br class=""></div><div class=""><a href="http://midas3.kitware.com/midas/download/item/317119/Hands_on_with_MAP-Tk.pdf" target="_blank" class="">http://midas3.kitware.com/midas/download/item/317119/Hands_on_with_MAP-Tk.pdf</a></div><div class=""><br class=""></div><div class="">FYI, MAP-Tk is (for the moment) a bit specialized to aerial imagery, so it might not work well out of the box on arbitrary images.</div><div class=""><br class=""></div><div class="">Good Luck,</div><div class="">Matt</div><div class=""><div class=""><div class=""><br class=""></div><div class=""><br class=""></div><br class=""><blockquote type="cite" class=""><div class=""><div dir="ltr" class=""><div class=""><br class=""></div><div class="">msr</div></div><div class="gmail_extra"><br class=""><div class="gmail_quote">On Tue, Oct 6, 2015 at 10:36 AM, Matthew Leotta <span dir="ltr" class=""><<a href="mailto:matt.leotta@kitware.com" target="_blank" class="">matt.leotta@kitware.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word" class="">Michael,<div class=""><br class=""></div><div class="">Thank you for reporting this issue. I’ll be upfront and warn you that I’m not a Windows developer and MAP-Tk is currently better tested on Linux and Mac. That said, we are working toward getting things better running on Windows more reliably.</div><div class=""><br class=""></div><div class="">In this case, the issues are related to Eigen byte-alignment issues as explained on the Eigen website linked from your error message:</div><div class=""><br class=""></div><div class=""><a href="http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html" target="_blank" class="">http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html</a></div><div class=""><br class=""></div><div class="">I have read this page before, and expected I would need to make changes at some point to work around these issues, but until now I had not come across a compiler that exhibited these symptoms. We should probably try to get a VisualStudio 2012 build up on our dashboard so I can work through those issues properly. I expect that will take some time. A shorter term solution is to try to disable alignment in your build. Can you try the solutions described at the bottom of the page at the above URL? In the "I don't care about vectorization, how do I get rid of that stuff?” section it suggests defining some macros to disable alignment. Let me know if that works.</div><div class=""><br class=""></div><div class="">In the meantime, I can at least work around that pass-by-value build issue. That function is pass-by-value because we intentionally want to make a local copy of the camera object to modify. However, I can explicitly make that copy inside the function instead. I’ve done this on the release branch of MAP-Tk and made a pull request (<a href="https://github.com/Kitware/maptk/pull/88" target="_blank" class="">https://github.com/Kitware/maptk/pull/88</a>). Can you try to build this branch and verify that it at least builds for you?</div><div class=""><br class=""></div><div class="">git fetch origin pull/88/head:dev/camera-pass-by-reference</div><div class="">git checkout dev/camera-pass-by-reference</div><div class=""><br class=""></div><div class="">If that works, I’ll merge this change into the release and master branches. Also let me know if the solution on the Eigen page helps.</div><div class=""><br class=""></div><div class="">—Matt</div><div class=""><br class=""></div><div class=""><br class=""><div class=""><blockquote type="cite" class=""><div class=""><div class=""><div class="">On Oct 5, 2015, at 1:24 PM, Michael Rosen <<a href="mailto:michael.rosen@gmail.com" target="_blank" class="">michael.rosen@gmail.com</a>> wrote:</div><br class=""></div></div><div class=""><div class=""><div class=""><div dir="ltr" class=""><div class="">Hello Kwiver-folk,</div><div class=""><br class=""></div>I'm building MapTK on Windows 7/64 using VStudio 12 and having two immediate difficulties: a compilation problem and a runtime assert.<div class=""><br class=""></div><div class="">MapTK is version 0.6.0, Eigen is 3.2.6<br class=""><div class=""><br class=""></div><div class=""><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">I've worked through getting Boost and Eigen built. When I compile maptk, the first problem is this:</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> [ 98%] Building CXX object tools/CMakeFiles/maptk_pos2krtd.dir/pos2krtd.cxx.obj</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> pos2krtd.cxx</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> C:\dev\maptk\tools\pos2krtd.cxx(220) : error C2719: 'base_camera': formal parameter with __declspec(align('16')) won't</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> be aligned</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> C:\dev\maptk\tools\pos2krtd.cxx(238) : error C2719: 'base_camera': formal parameter with __declspec(align('16')) won't</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> …</span></p><div class=""><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> </span><br class=""></div><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">I fixed these compilation errors by changing the signature of the offending function from pass-by-value to pass-by-reference:</span></p><div class=""><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> </span><br class=""></div><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> /// Convert a POS file to a KRTD file</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> bool convert_pos2krtd(const maptk::path_t& pos_filename,</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> const maptk::path_t& krtd_filename,</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> maptk::local_geo_cs& cs,</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> maptk::camera_d& base_camera, // msr. was pass-by-value</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> maptk::rotation_d const& ins_rot_offset = maptk::rotation_d())</span></p><div class=""><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> </span><br class=""></div><div class=""><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> </span><br class=""></div><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">That allows everything to compile and link. The next problem was that I got an Eigen Assert when I run any of the executables:</span></p><div class=""><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> </span><br class=""></div><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">C:\Program Files (x86)\MAPTK\bin>maptk_analyze_tracks.exe</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">...</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">Assertion failed: (reinterpret_cast<size_t>(array) & 0xf) == 0 && "this assertion is explained here: " "<a href="http://eigen.tuxfamily.org/d" target="_blank" class=""><span style="color:purple" class="">http://eigen.tuxfamily.org/d</span></a></span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">ox-devel/group__TopicUnalignedArrayAssert.html" " **** READ THIS WEB PAGE !!! ****", file c:\program files (x86)\eigen\include\eigen</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">3\eigen\src/Core/DenseStorage.h, line 86</span></p><div class=""><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> </span><br class=""></div><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class="">The web page says that classes which contain certain Eigen data structures as members need to use the</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""> EIGEN_MAKE_ALIGNED_OPERATOR_NEW macro to override the "new" operator so that those members will be 16-byte aligned. I've done that but am still seeing the assert.</span></p><p class="MsoNormal"><span style="font-size:11pt;font-family:Calibri,sans-serif" class=""><br class=""></span></p><p class="MsoNormal"><font face="Calibri, sans-serif" class=""><span style="font-size:14.6667px" class="">Can anyone offer insight into any of this?</span></font></p><p class="MsoNormal"><font face="Calibri, sans-serif" class=""><span style="font-size:14.6667px" class=""><br class=""></span></font></p><p class="MsoNormal"><font face="Calibri, sans-serif" class=""><span style="font-size:14.6667px" class="">msr</span></font></p><p class="MsoNormal"><font face="Calibri, sans-serif" class=""><span style="font-size:14.6667px" class=""><br class=""></span></font></p></div></div></div></div></div>
_______________________________________________<br class="">Kwiver-users mailing list<br class=""><a href="mailto:Kwiver-users@public.kitware.com" target="_blank" class="">Kwiver-users@public.kitware.com</a><br class=""><a href="http://public.kitware.com/mailman/listinfo/kwiver-users" target="_blank" class="">http://public.kitware.com/mailman/listinfo/kwiver-users</a><br class=""></div></blockquote></div><br class=""></div></div></blockquote></div><br class=""></div>
</div></blockquote></div></div></div><br class=""></div></blockquote></div><br class=""></div>
</div></div></blockquote></div><br class=""></div>
</div></blockquote></div><br class=""></div></div></div></div></blockquote></div><br class=""></div>
</div></blockquote></div><br class=""></div></div></div></div></div></blockquote></div><br class=""></div>
</div></blockquote></div><br class=""></div></div></div></div></div></blockquote></div><br class=""></div>
</div></blockquote></div><br class=""></body></html>