Alexandre Bizeau's profile - activity

Have you found something ? Because I'm having the same issue right now. I'm trying to get a better disparity map too, but it really difficult.

Really strange... Try to update the firmware/hardware, sometime they release patch. The camera just look to have a problem.

Can you post an example of the depth you have ? Because this code seems to be really good and me I don't have any trouble with near mode.

I use OpenCV 2.4.6 to rectify my stereo system.

The parameter matrix obtained after the stereoCalibrate and stereoRectify show that my system is vertical stereo, so the epipole line are vertical.

I don't know if that change something for the stereoBM/stereoSGBM algorithms but I don't have a good disparity map. I want to know what can I do to expect to have a nice disparity map with some uniform region.

My image are high-resolution 3000x4000 and well rectify. But when using stereoBM/SGBM to obtain my disparity map, the result are bad.So when I reproject, I have a bad point cloud. (I'm using PCL pcd_viewer). When I do stereoCalibrate, d = 4.15, so I think that mean it's the average pixel error of my calibration.

So their my result, this is an example of rectification.

And this is an example of the disparity map obtain after that, so when I use reprojectImageTo3D(), I have a pretty bad cloud that look like a prism.

So the question is: How can I use stereoBM/SGBM with vertical system to get a good result on disparity map and after be able to reproject it to get the cloud ?

I was thinking about that too. I still on the point to get in the code and try to understand it before doing something.

This is not really a normal question here, but I would like to know :

What the best way to modify source code ? And can I have advices on how to do it ?

1 - Is it better to change code in source and build OpenCV as write in this tutorial.

2- Or it easier to create a new class and just rename function to don't have conflict.

The problem is if I use the (1) when I'll use the next OpenCV version, I'll need to re-build it again so maybe it not the best way.

And for (2), the problem is more about the include that can be a really hard thing to deal with.

And I want to maybe use this modification soon, so that why I don't want to ask as a new feature for next OpenCV version.

I would like to implement the "probabilities estimates" into SVM.

As explain here.

@StevenPuttemans I went to OpenCV code and figured out the implementation has been create for double and float. And using the function JacobiImpl_ in modules\core\src\Lacpack.cpp. So I don't think the problem is that. But maybe @Philipp Wagner is right because the function start with eps = std::numeric_limits<>::epsilon(). But C++ EPS is equal to MatLab. I still working on this. Ty

@Constantin Like Steven post as answer, you have to change Lib in linker. Because there Debug and release compiled lib. This is the only thing I see. Else provide more information.

You didn't provide that much information. Have look at your project Include Directories with good Configuration (Release) and Platform ? Same for your Library Directories and into linker Input ?

I tought my accuracy problem were maybe about something like that ( Condition number). I read the article you give me and I try it. So my 100x100 is K. It the one I need to do rcond on it right ? If yes, rcond result is really bad : 6.8180e-020 and my MatLab eps is : 2.2.04e-016, so I'm really near 0.0 then my matrix(K) is really ill-conditioned. So what can I do ? I will continue to read about it because it can be a good point and maybe this is why I have sign problem and same for accuracy. Working on it ! Thanks you.

@Philipp Wagner : Thank you for your time. Right now, I have found some implementation of eig in MatLab in some toolbox and I just trying to see if there a lot of difference between the built-in and those function. Stay alert, I'll post an edit on my answer with more details about my result and every thing.

I tried the SVD function and in same time, I tried PCA one (Suggested Eigen reference). My result for SVD are same as Eigen losing accuracy and sign problem. With PCA, the result are really not matching and looks more bad than my float with eigen. Both function were a good idea to try but they are not the answer to my problem. Thanks Philipp.

I won't be able to found the Matlab eig function because it's a Built-in function. And I can't be sure the octave one can match. But I will try to work on that. And maybe try the SVD like Philipp said. Thanks.

Thanks Philipp, the problem is because I need to get Eigen Vector. I know with eigen values, I can obtain the vector but more processing. But can be an idea to try it.

And I never tried to used float in Matlab, but if you think it can be a good, I can try it now.

When I'm using float, I'm facing the same problem, sign problem with the eigen vector and bad precission on the eigen value. But It start losing accuracy at like value 9 instead of 25-30 and at value 100 with double I have a difference of 0.000001 and I have 0.01 with float. And Double - Float - Matlab have different sign result. Double and Float sign not matching either.

Hello,

I need to retrieve the eigen vector from a matrix. The problem is I don't match the result same as my matlab code.

I have a matrix symmetric 100x100 and trying to obtain eigen value and eigen vector from it. I work with double matrix (CV_64F) to have the best precision as possible (Already tried Float and it fail more).

My eigen value seems good but the vector loses some accuracy for each value ( Ex : value 1 to 25 exactly match with matlab but further you get to 100 , more I am losing precision, but I can work with that.)

But the problem is more with the eigen vector. The result is same as eigen value, lose precision but the problem is with the sign. If I take the firsts 25x25 results. I'm totally matching with matlab but randomly have positive or negative value. So I got wrong information at the end.

Right now I'm using cv:eigen function like this :

cv::eigen(oResultMax,oMatValue,oMatVector);

I have already tried the SelfAdjointSolver from the library Eigen. Eigen SelfAdjointSolver

Anyone have an idea ? Or suggest me something ?

EDIT : (Add image of result)

Here you can see the image of the eigen value. If you refere to the MatLab result, the double result are almost same till the 27th and for the float till 14th.

And those Image, you have Eigen Vector in order : MatLab , OpenCV Double and OpenCV Float.

As you can see double matching until 35th and float until 15th. And the sign of the result are differente from Matlab, Double and Float

@steven : This is what I want to do, but before, I just ask :) Thanks you.

I search into svm documentation.

I looking for a parameter to configure to print every step of the train_auto during cross-validation. Libsvm have it. It's -q for quiet mode. I want to know every step which C and Gamma I'm testing and what the Accuracy of this test at the end.

Is there any function already implemented for that ? Because right now, I can only print the C and Gamma at the end of train_auto.

Thanks you in advance.

Just in case, I haven't read the Java documentation yet. But just beware about BGR channels ordering if you are trying to use it.