dohnjoe's profile - activity

Look refer to tetragramm's answer below, he puts it plainly. RAW is when do ZERO compression to the data before you save it (usually). You save it as it as. JPEG and PNG are compression methods in which you selectively choose certain kinds of values and throw the rest of them away, which you would have known, had you looked up image compression methods as I had asked you to.

I hope you do realize the contradiction in your own statement. You yourself brought in jpeg and raw into the mix when there was no need to

"specific format like jpeg or png".

I answered in the context in which you asked the question in, if you want an answer to something else, you should ask accordingly.

As I said, next time please frame the question with a specific goal of what you actually want.

Really? Even if I declare the svm model once and then sequentially autotrain using N+M samples? Wont the parameters and support vectors learned from N live on and be used as an initial starting point for the M samples? (because I dont delete the original svm model itself, so there must be some persistance). Can you show me the code to back this up, that the SVM model itself gets erased?

Have you tried varying the linewidth and font size?

From what I could put together from your un-indented code. It seems you're just banging together two images and expecting them to "seamlessly" flow into one image.

It would be better I think if you post your code about the laplacian gaussian instead of this one or at least posting a link of where ever you obtained the code from. Otherwise I would suggest looking at the cv::Stitcher class for automatic image stitching.

I dont quite understand what you're trying to ask here but I hope you do realize that PNG, JPEG and RAW refer to the image compression methods (basically a turn on how you store images) and RGB values refer to the colorspace in which images are saved. An image could be saved in any format and it need not be RGB all the time (well but it technically is due to RGB being a common standard).

If you're asking about the differences between PNG, JPEG, and RAW. Then I would suggest reading up on image compression

When VideoCapture procures a frame from a video of a file then the answer to your question would be "what is format of the video file?" moreover "what is the compression method used for the video file?" I would suggest reading up on video compression formats / video codecs.

I would suggest framing the question with a clear understanding as to what it is you want to ask about would always be a good start.

So it is common knowledge that due to a bug in OpenCV when you try to load a non-linear kernel SVM after saving it you get an error.

Refer here for more: https://github.com/opencv/opencv/issu...

People HAVE gotten the SVM to load but were unable to obtain the same result (classifier data was corrupted after loading). I tried testing this by writing my own svmloader function.

Ptr<SVM> svmloader(const String& filepath)
{
    FileStorage fs;
    fs.open(filepath, FileStorage::READ);

    Ptr<SVM> svm = Algorithm::read<SVM>(fs.getFirstTopLevelNode());
    return svm;
}

This works but I can as of now confirm that I've tested model pre and post loading and the results seem worse post-loading. That is after you train a classifier and you test it, it will be fine but if you save it and load it again it will get effed up.

So I think I want to use C++'s Object serialization to solve this problem. How can I write arbitary objects to a file and then read from that file?

Also can someone confirm if the above bug has been fixed in OpenCV 3.2.0 ,I COULD NOT FIND ANYTHING IN THE RELEASE NOTES

Any help is GREATLY APPRECIATED folks

I want to use SVM.autoTrain in OpenCV.

Say I have N samples, I train an SVM model using these N samples using cv::SVM::autoTrain().

Say I have another M samples of data.

I used the same above trained model with cv::SVM::autoTrain to train these M samples.

Will this overwrite the previously trained model (trained with N samples) with one created now with M samples?

OR

Will this augment the previously created model (with N samples) with the M samples? (That is, will the model created in this case be equivalent to the case if the M+N samples were part of the same dataset fed to the SVM together?)

There is very good documentation about cv::SVM but not anything that elaborates on this issue exactly.

Thanks