2020-12-08 13:21:53 -0600 | received badge | ● Notable Question (source) |
2020-10-28 04:31:15 -0600 | received badge | ● Nice Question (source) |
2020-06-11 03:06:51 -0600 | received badge | ● Popular Question (source) |
2020-03-10 09:37:59 -0600 | received badge | ● Notable Question (source) |
2019-05-29 13:37:58 -0600 | received badge | ● Popular Question (source) |
2019-03-02 05:34:56 -0600 | received badge | ● Famous Question (source) |
2018-02-05 04:40:17 -0600 | received badge | ● Notable Question (source) |
2017-10-28 10:35:53 -0600 | received badge | ● Popular Question (source) |
2017-04-13 07:24:42 -0600 | marked best answer | Creating custom kernels I want to create custom filters using filter2d function but most of the examples I see use custom methods like ones,zeros and eye but I want to create a 3x3 matrix which contains the values I want. Like for e.g, [ 1 4 6 -1 3 5 -1 -2 2 ] How do I make such a kernel using filter2d function? |
2016-12-28 12:29:26 -0600 | asked a question | Can't build opencv 2.4.11 with python When I try to build opencv 2.4.11 with both c++ and python, an error comes up for python- The thing is numpy is installed for python 2.7.6 and not for python 2.7.9 which is what opencv is trying to build from. How do I rectify this issue? How can I make opencv build particularly with python 2.7.6 or install numpy with python 2.7.9? |
2016-12-12 21:38:07 -0600 | commented question | Opencv 3 Dnn caffe_googlenet Hey leofs, has your error been solved? Look at usr/local/include/opencv2 path and see whether your dnn folder is there. |
2016-12-12 00:50:51 -0600 | asked a question | Floating point exception- Running sample program dnn module I am using Opencv 3.1.0 on ubuntu 14.04. I have successfully built Opencv and wanted to try reading a caffe model and classifying. I used the program and models from this tutorial I am able to compile the program but running it gives me the error below- The floating point exception is coming from line 109 in net.forward() method. I am not able to understand why this problem is coming up? |
2016-12-11 23:40:05 -0600 | commented question | Can't run sample dnn program Sorry guys, it was a silly mistake. There was a space in the beginning of the caffemodel file name, so it couldn't read it. Later I got another error about failing to parse NetParameter file, so I had to download the caffemodel again and it got solved. Now I have another error, so I will spend some time to see if there is a similar error online, or else I will put it as a separate issue. |
2016-12-10 02:04:43 -0600 | commented question | Can't run sample dnn program Hi break, I actually put all the models and image in the same folder with the program and still the error comes. |
2016-12-09 23:45:17 -0600 | asked a question | Can't run sample dnn program I am using Opencv 3.1.0 on ubuntu 14.04. I have successfully built Opencv and wanted to try reading a caffe model and classifying. I used the program and models from this tutorial I am able to compile the program but running it gives me the error below- I don't understand why it can't read the models? |
2016-12-09 22:32:00 -0600 | commented answer | Trying to run dnn module Hey pi, using cmake on command line, I got a linking error for protobuf. So I manually copied protobuf from their github repo and it built perfectly fine. I can compile the sample dnn program but can't run it. I guess I will ask it as a separate question. |
2016-12-09 22:31:47 -0600 | commented answer | Trying to run dnn module Hey pi, using cmake on command line, I got a linking error for protobuf. So I manually copied protobuf from their github repo and it built perfectly fine. I can compile the sample dnn program but can't run it. I guess I will as it as a separate question. |
2016-12-09 10:47:17 -0600 | commented question | Trying to run dnn module Hey berak, I just copied the protobuf folder from github and it built properly. I am able to compile the sample program but not able to run it. If it doesn't work, I guess I'll just ask it as another question :) So I'll raise it as an issue on github right as the files should be automatically downloaded instead of manual search. |
2016-12-09 04:05:20 -0600 | commented question | Trying to run dnn module I am getting cmake warnings which say that the protobuf files are missing which CMakeLists.txt should have downloaded. Will I have to manually move all these files along with the folder to the directory or is there another solution? |
2016-12-09 01:25:10 -0600 | commented question | Trying to run dnn module Hey berak I'm having such an awesome time here with users like you who help :) . Thanks for the links berak, you might want to edit the second and third links as it is leading to resource not found. For second link in url path, the file name should be boostdesc_bgm_bi.i instead of boostdesc_bgm.bi.i . For third link in url path, the file name should be boostdesc_bgm_hd.i instead of boostdesc_bgm.hd.i . Now I'm facing this Cmake error sadly :( Do you know how to resolve this? I need libprotobuf to read caffe files. |
2016-12-08 23:09:10 -0600 | commented question | Trying to run dnn module Hey berak, I tried following this cmake command from this website It seems to include the Opencv_contrib module but I'm facing this failure of download of boostdesc_bgm_bi.i file. I tried finding this file online but to no avail and this stackoverflow question faces the same problem but it hasn't been answered since Novemeber 19th. This is the cmake error output- Failed to download boostdesc_bgm_bi.i. Status=1;"Unsupported protocol" |
2016-12-08 23:09:01 -0600 | commented question | Trying to run dnn module Hey berak, I tried following this cmake command from this website It seems to include the Opencv_contrib module but I'm facing this failure of download of boostdesc_bgm_bi.i file. I tried finding this file online but to no avail and this stackoverflow question faces the same problem but it hasn't been answered since Novemeber 19th? This is the cmake error output- CMake Error at /home/kevmepls/opt/opencv/cmake/OpenCVUtils.cmake:1025 (message): Failed to download boostdesc_bgm_bi.i. Status=1;"Unsupported protocol" Call Stack (most recent call first): |
2016-12-08 11:31:03 -0600 | commented question | Trying to run dnn module Even with absolute path, the cmake output is the same. |
2016-12-08 10:56:29 -0600 | commented question | Trying to run dnn module Could it be a problem with cmake then? Is there any other reason why it may be failing? |
2016-12-08 10:42:02 -0600 | commented question | Trying to run dnn module Yes, it is the correct path. Sometimes without the |
2016-12-08 10:41:18 -0600 | commented question | Trying to run dnn module what did you do to format the output? Also as you can see from output, the extra modules from opencv_contrib are not being built but are also not unavailable? How is that possible even when my cmake path points to the opencv_contrib folder? |
2016-12-08 10:32:31 -0600 | commented question | Trying to run dnn module I'm really sorry but I'm not sure how to edit the cmake output properly in opencv q&A |
2016-12-08 09:37:03 -0600 | commented question | Trying to run dnn module Hi berak,
|
2016-12-08 06:28:22 -0600 | asked a question | Trying to run dnn module I have installed Opencv 3.1.0 along with Opencv_contrib module with the aim of reading caffe models into Opencv. I followed the official Opencv installation documentation and with a change in Cmake command, I was able to build it successfully on Ubuntu 14.04lts. When I type the following command to run the sample program - I get the error below. I looked at the opencv_contrib github repo and they mentioned that you have to run it with a linker to the specific module which you want. So my command then became- But I still get the same error. Does this mean that my dnn module hasn't been built properly? The error is from the header file opencv2/dnn.hpp not being recognised but I don't know why? The sample program is from this link- sample program to load caffe model and run Output of cmake- (more) |
2016-11-07 23:33:36 -0600 | commented question | Sobel edge detection Implementation I do padding to take care of edge pixels. If I use gradient_y.at<char>(r - radius, c - radius) = s; then my edges become much worse. I don't understand the reason why but when I'm dividing by 30, the edges becomes much more clearer. In the link below, final1.png is s/30, final2.png is s. |
2016-11-06 23:39:05 -0600 | commented question | Sobel edge detection Implementation Hey LBerger, actually that indexing is required to access all the pixels in the image. Initially, I pad the image with one pixel across the whole border so that I could implement correlaton properly. I'm calculating gradient magnitude which is sum of square of gradients in x and y direction, so the negative values are getting squared. Or am I suppose to ignore the negative gradients? Oh so sobel doens't require a threshold? Do you have any idea how the OpenCV implementation works as I'm almost following allthe steps in Sobel edge detection but still my output is different from Opencv? |
2016-11-05 13:33:54 -0600 | asked a question | Sobel edge detection Implementation I am trying to implement sobel edge detection from scratch but my output can't seem to match with OpenCV's sobel function. I performed correlation on the image with the sobel operator in both x and y directions and then computed gradient magnitude as square root of sum of squares of magnitudes in both x & y direction. I believe the problem is how I assign the threshold for edge detection. Images- 1.Original Image- 2.My implementation of Sobel Edge Detection - 3.Opencv Sobel edge function output - Code- (more) |
2016-11-03 19:49:17 -0600 | commented answer | Canny Edge Detection- Non Maximum Surpression Implementation Yes, it actually gave me much thinner edges. Do you have any idea why my sobel gradients isn't coming out properly? I changed the value of my sobel operator in y direction to int sobel_y[3][3] = {-1,-2,-1,0,0,0,1,2,1}; . I thought that was the mistake but barely any change is there. |
2016-11-01 04:31:21 -0600 | asked a question | Canny Edge Detection- Non Maximum Surpression Implementation I am trying to implement the Canny edge detection algorithm from scratch with the help of OpenCV. I am facing a problem implementing the Non-Maximum Suppression step which helps to thin the edges. My logic is to first compute the intensity gradient vector, then group it in either 0,45,90,135 degrees direction and then try to find local maxima. The method to find this local maxima is by making sure that current pixel is greater than succeeding and preceeding pixel in same direction. If not, I assign value of zero to this pixel. Using this logic, I'm still not able to thin the edges. I feel the error is when I'm computing the intensity gradient vector for each pixel. Here is my code-
(more) |
2016-10-18 11:06:39 -0600 | commented question | Correlation Implementation Sorry for the late reply, @LBerger I didn't understand exactly how template matching can be used for correlation implementation. |
2016-10-13 03:55:58 -0600 | asked a question | Correlation Implementation I want to basically write code from scratch to do correlation on a given image with a kernel(without using filter2d function). I am trying to find a general way of correlating image with kernel( mxm size where m an is odd number) but I can't seem to think of a good algorithm that will work for kernel of all sizes. In case of a boundary issue, we compute correlation for each pixel of image as long as the center of the kernel is on this pixel. My idea is to first classify the pixels into two types, first one being pixels where simple multiplication with kernel is fine. The problem comes with the second type that is the boundary pixels, the pixels at the edges of the image. I can't seem to find a general way and I seem to end up hard-coding it to compute correlation for these pixels for different sizes of kernel. Does anyone have any ideas or suggestions for a simpler implementation? |
2016-07-26 21:51:58 -0600 | asked a question | findChessboardCorners returing false boolean value? I am using findChessboardCorners function to find the corners in a chessboard image. When I am executing the function on this image, I am getting an array of detected corners but the bool value returned by the function is zero. Shouldn't it be one since I'm getting an array of detected corners? Code- |
2016-07-23 03:11:25 -0600 | commented answer | why composeRT function is showing error for column vectors? Thanks a lot berak for the swift answers |