Steve7's profile - activity

Hi I am trying to understand various parameters that HoG uses but it isn't documented properly in the docs (CPU version). There are some explanation on the GPU::HogDescriptor though but I find it difficult to understand as I was new to the whole image processing area but I needed to implement a mini-project using HoG asap.

So my question is, how does these parameters (win_size, block_size, block_stride, hit_threshold, win_stride, padding, group_threshold etc.) affects accuracy, performance, reliability, false positives.

By the way I did try to read the Dalal,Triggs papers but there are too many concepts to understand but too few amount of time. Therefore I was wondering can anyone shed some light on how these parameters affects those attributes. Thanks a lot

Yeah I was suspecting this was the case as I have mentioned cloning solves the issue. However, thanks for confirming my suspicion, at first I don't know why it was designed this way, I guess you have a point on the performance issue.

Yeah I did include the initialization but I simplified the code when I posted it here. The problem persists even with the initialization though.

Hi I am fairly new to OpenCV and recently I encountered one really weird situation when I tested out cv::absdiff on a video feed (I want to compare the first frame of the video with each following frame).

VideoCapture capture("768x576.avi");

Mat first_frame;
Mat current_frame;
//Mat diff;

capture>>first_frame;

while( (char)keyboard != 'q' && (char)keyboard != 27 ){

    capture >> current_frame;
    //cv::absdiff(current_frame, first_frame, diff);

    imshow("first_frame", first_frame);
    imshow("current_frame", current_frame);

    keyboard = waitKey( 30 );
}

The weird thing I encountered is, you can see I only modified Mat first_frame once (putting in the first frame which isn't a black frame) but when I imshow both first_frame and current_frame on the while loop, they both "play" like a normal video which is weird since I do not modified first_frame in the while loop. Can anyone please explain why is this happening?

P/S: I solved it by cloning the first_frame into another Mat by still, why is this happening?

In fact this is one of the reasons I chose not to use a classifier, the efforts in locating "good" positive samples and negative samples. Anyway thanks for the suggestion, I'll try it if I have more time.

@StevenPuttemans Oh I didn't realize it would take such short time as from what I read it requires hours, even days to train a classifier. However would you mind pointing me a direction to, perhaps a tutorial on training a LBP cascade in a short time? Thanks a lot.

Can you provide some code examples on how to retrieve the number of outliers/inliers from the output mask? What should be the data type of the said output mask?

Hi I tried out your code but this line Mat H = findHomography( obj, scene, CV_RANSAC, 3, mask ); gives error "OpenCV Error: Assertion failed (mtype == type0 || (CV_MAT_CN(mtype) == CV_MAT_CN(type0) && ((1 << type0) & fixedDepthMask) != 0)) in create, file C:/slave/builds/WinInstallerMegaPack/src/opencv/modules/core/src/matrix.cpp, line 1486".

So I swapped char with uchar and the error goes away, am I supposed to do this?

Hi thanks for the reply. I should've mentioned I am really new to OpenCV and I don't really understand your approach, would you mind giving me some code examples?

Hi I was trying to detect an object in a video feed. So far I have explored Template Matching but the results wasn't satisfactory as the video feed may or may not contain the object and is subjected to rotation and differing scale. Furthermore I have no time to train a cascade classifier or sorts, therefore I have stumbled upon SURF+homography approach from this link: http://docs.opencv.org/doc/tutorials/features2d/feature_homography/feature_homography.html.

The results was good but what I wanted to "get" from my program is simply an answer to "is the object currently present in the video?" (Present/Absent). So how can I get this answer using the SURF+homography approach?