mvuori's profile - activity

You have a generic, very basic string building question that has nothing to do with OpenCV and has no place here.

You must have read the FAQ and noticed that e.g.the following questions are not allowed: "Questions about your homework

(Here your code is completely illegibly formatted and impossible to even try to understand.)

Well, you are passing that function a 16 bit image, while it tells you that it only accepts 8 or 32 bits. So, after lapl

This is because an YUVY image height and width must be a multiple of 4.

You can't make it work for anything as long as you use single backslashes in file paths

Impossible to say, as FileVideoStream is not an OpenCv class (but at least you need to check what read() returns)

...Your code doesn't output the "Found 3 faces in Input image #2!" message so it comes from another piece of code... B

I think you have installed hdf5 runtime, but lack the development package libhdf5-dev

(I can't see any circles in the image, do it might be hard for OpenCv too...)

That's because you lack permissions in that special template folder. You are running into Windows' features, not OpenCv'

I guess you should forget samples.findFile and just call imread with your own path to any file you wish to open, unless

Guessing what is wrong would be easier if there was code to guess from... but the error implies that something is too cl

When you googled the warning message, you must have noticed that LOTS of people have encountered that warning and discus

... and mark the circles that aren't detected...

(Of course, a program should not crash just due to delays or number of threads. Delays can be handled and threads orhest

Your code doesn't do any checking whether the function calls succeed and what the variables contain... Start with that.

You might want to remove the "a few libraries", then add them one at the time and identify any changes in your program's

That is not "a mask", but a variable named mask. What it represents and what it is used for, and how, depends on the pro

If the system has a cache, the problem is that your hard drive becomes the bottleneck aften the cache fills up. Obviousl

I don't understand what the code tries to do - which is the first thing to describe and for others to understand.. The

I guess there are 2 alternatives: 1) modifying the software and hoping it works... 2) as only some videos are too large,

First two advices are about asking help: 1) If "it says some VideoCodec_RGB24 error", report what the error exactly is,

8S is really a char, but you are feeding the Mat a buffer of ints - which in 32 bit program are 32Ss...

A minute of googling could have made you find out that you can use getWindowProperty() to check whether the window is ok

If by "raw" you mean a bayer "image", you can't, and that would be a silly thing to do

If it worked a couple of hours ago, but not now, something has changed - perhaps crashed. It is not uncommon to debug a

I don't know, but while it freezes on that line, things may have failed elsewhere. I'd try without the yield command

I _think_ that is expected, especially if the pinhole is an attachment over the device's autofocus optics. The device's

One possibility is that it has something to do with your code - which you don't show

You don't even test that videcapture. read() succeeds... that's what ret is for. Likely your videocaoture is not even op

There has never been. version 3.6.1 of Android.

The problem is in the code that you, for some reason, don't show: your function getOrderedPoints gets the points wrong.

Googling shows that the code might imply stack overflow, but also pethaps running out of consecutive memory. So the prob

The corner blobs are the ones whose distance from the corner is smallest. Calculate that for each blob, for each corner.

If by intensity you mean luminance / lightness, you can split() a Lab image into its components and thus get the L compo

"The good people of arduino" don't seem to have much to do with OpenCV...

Did you test the images in an image editor that they really are the same? You could compare an image with itself. Two ca

Nobody can help you here, as your code doesn't use OpenCV for the detection - just like it is explained in the code. You

Just put your current frame into a variable named "previous_frame" and once you read the next frame, you have it there

It would be easier to see what you do wrong if you tell what happens wrong...

You should learn about debugging, that's the thing programmers do to find where and why code crashes with the data they

No known OS uses double forward slashes as path separator.

Anything that is done to a still image is trivial to do to every video frame, for anyone that can program and has spent

Don't worry about VideoCapture. What you seem to need is jut a read() that return an image by some logic. Perhaps just g

The "sometimes" errors are a clue. Your code fails sometimes because there are errors somewhere. I'd start putting check

... Just remove things until the leak stops (don't care about how it works when hunting the bug). (Note that it is a g

"it doesn't work" doesn't explain anything about the problem. You need to tell in detail how it doesn't work: what you e

Measuring: you could shoot something that shows an exact time and compare that to the time you retrieve a frame that sho

In case OpenCV doesn't offer a function for this, justt use basic trigonometric calculation to find out if distance from