2020-11-30 00:56:01 -0600 | marked best answer | SVM bias on weights of positives and negatives For the purpose of algorithm comparison I want to compare the detection result of my cascade classifier trained using the viola & jones framework versus an SVM HOG classifier I have trained. However, the VJ framework has no problem with an uneven ratio for positives and negatives, for example 500 positives and 100 negatives (don't tell me this is wrong please, my whole PhD is about proving that context awareness can reduce the negative numbers effectively. . If I would like to convert this to SVM classification however, I am reading at many blogposts that using CvSVM should be done with an equal number of positive and negative training samples. If not I should add a bias, but however, I cannot find an example that does exactly that. Could anyone give me some directions on how I should apply this? Is this even possible within the OpenCV framework? Kind regards, Steven EDIT 1: I have noticed people suggest using libSVM or SVMlight for training the model. However, I would like to stick to openCV only, which should actually be possible. |
2020-11-17 17:23:53 -0600 | received badge | ● Nice Answer (source) |
2020-11-16 14:54:24 -0600 | marked best answer | Passing multiple parameters with the setMouseCallback function Is it possible to do? As of know I am using the following to pass an image to the callback: But I would like to add a second pointer to a vector of points to keep storing them also, just don't know how to add multiple parameters. Anyone has any idea without creating global variables? |
2020-10-28 03:13:58 -0600 | received badge | ● Great Answer (source) |
2020-10-28 03:13:58 -0600 | received badge | ● Guru (source) |
2020-10-20 09:27:39 -0600 | marked best answer | What is the average training time needed for LBP object detector to reach a next stage? I see many topics appearing about cascade training for object detection. However, what I do not seem to find are discussions about how fluently training should actually be. I am training a LBP object detector using the boosted cascade of weak classifiers, done by using the train_cascade algorithm. The detector is used for detecting cars in aerial imagery. I have used 200 positive examples and 1000 negative examples. I asked to train 35 stages, which the trainer is actually doing, but when I check how fast the algorithm is processing the negative samples, I see weird results. At first stages processing goes very fast. At stage 19 however, processing goes about a sample each 5 - 10 minutes. Just wondering, since LBP has been reported as beïng trainable in matter of a day, where HAAR needs a week, how fast training goes with others. Also, I have the idea that assigning more memory to the process, by increasing the precalcValBufSize and precalcIdxBufSize parameter to 2048 MB each, doesn't seem to influence the training more, than actually assigning 500 MB to each. Any suggestions there? UPDATE I have done multiple tests on this, still not being able to define a relation between LBP training time and the algorithm itself. Still open for suggestions! |
2020-09-13 08:20:27 -0600 | received badge | ● Popular Question (source) |
2020-06-10 21:03:17 -0600 | marked best answer | cvSetMouseCallback function pointer WinForms I got a function Then the function And it doesn't seem to work using the function name as function pointer, which works perfectly in a CLI application, but not in this WinForm application header file. Anyone could help? |
2019-09-04 02:58:22 -0600 | edited answer | traincascade's error (Required leaf false alarm rate achieved. Branch training terminated.) This is not an error! Given the samples presented and the settings of the training, your cascade has reached its desired |
2019-09-04 02:56:58 -0600 | commented answer | traincascade's error (Required leaf false alarm rate achieved. Branch training terminated.) You are hitting the nail there ... since it is indeed the max allowed FArate, it should be lower to be more strict, and |
2019-08-28 03:14:30 -0600 | commented question | The results of cv::eigen() are different on Win10 and Ubuntu. Exactly my point, on 2 OS systems, even reading from the exact same source, could be given different values. |
2019-08-28 01:54:09 -0600 | commented answer | OpenCV 4.1.0 not changing brightness The documentation is indeed not that clear. We might want to propose a pull request with a chance on that level. |
2019-08-23 08:08:37 -0600 | commented question | How to run OpenCV DNN ON GPU ANDROID Its just strange to copy paste someones question. If you have the same issue, then simply add a comment beneath the ques |
2019-08-23 08:00:42 -0600 | commented answer | OpenCV 4.1.0 not changing brightness My guess is the internal translation of the indexes is somehow broken ... could you maybe open up an issue at github wit |
2019-08-23 07:59:09 -0600 | commented question | Python - Facial Recognition Hahaha I am totally unrelated to nvidia :D Never did any work for them or with them. Deepfakes are not the solution howe |
2019-08-23 07:54:29 -0600 | commented question | prior_box_layer.cpp:242: error: (-215:Assertion failed) !params.has("step") in function 'cv::dnn::PriorBoxLayerImpl::PriorBoxLayerImpl' @NickKao, dont worry. I deleted them. Seems to be a system fault :) |
2019-08-23 07:53:50 -0600 | commented question | OpenCV 4.1.1 Imread() & Imwrite() Crashes Program Immediately [Qt] ? Then my guess is this issue is due to your Qt environment are wrongly configured and the issue is not within OpenCV itse |
2019-08-22 09:53:02 -0600 | commented question | prior_box_layer.cpp:242: error: (-215:Assertion failed) !params.has("step") in function 'cv::dnn::PriorBoxLayerImpl::PriorBoxLayerImpl' did not know that :D thanks for giving a heads up |
2019-08-22 09:51:08 -0600 | commented question | OpenCV 4.1.0 not changing brightness can you add between each read() operation a wait? I know that when grabbing frames in a loop, in C++ I always add a wait |
2019-08-22 09:29:30 -0600 | commented question | OpenCV 4.1.0 not changing brightness But it tells us what you are trying to change. You cannot expect us to dig into the code to figure it out. Are you sure |
2019-08-22 09:28:11 -0600 | commented question | Not able to identify the images correctly On top of that, there is this book opencv 3 blueprints, which has a very detailed discussion on the process as well as a |
2019-08-22 09:25:56 -0600 | commented question | OpenCV 4.1.1 Imread() & Imwrite() Crashes Program Immediately [Qt] ? You cannot just transfer dlls from one machine to another without having an exact copy. It is very tricky. I would urge |
2019-08-22 09:23:48 -0600 | commented question | How to run OpenCV DNN ON GPU ANDROID That is indeed weird. @markcowell23, be warned that it is not normal simply copy pasting others questions. If this keeps |
2019-08-22 08:00:51 -0600 | commented question | Python - Facial Recognition @holger, there is no way you can mimic facial microexpressions over time inherently specific to the person in question. |
2019-08-22 07:57:40 -0600 | commented question | prior_box_layer.cpp:242: error: (-215:Assertion failed) !params.has("step") in function 'cv::dnn::PriorBoxLayerImpl::PriorBoxLayerImpl' The duplicate posts have been deleted ... you posted 14 times the same question. In the future, please pay attention to |
2019-08-21 07:30:06 -0600 | commented question | error in cv::moments function As you can see in the moments tutorial you need a contour, which is a vector of point elements as input format. My guess |
2019-08-21 06:14:35 -0600 | commented question | opencv.js building WASM cv.face.createFacemarkLBF() Keep in mind that the facemark code was developed during GSoC2017 and is still since then in contrib module because of t |
2019-08-21 06:01:55 -0600 | commented question | OpenCV3.4 DNN forward custom and pre-trained Tensorflow if you really want quick help on DL issues, open up a full report on github repo and @dkurt will jump in quite suddenly |
2019-08-21 05:56:37 -0600 | commented question | How to run OpenCV DNN ON GPU ANDROID Prebuilt OpenCV has no OpenCL support I think ... but anyway, its always better to select all options yourself and build |
2019-08-21 05:53:47 -0600 | commented answer | Camera properties dialog (DirectShow) no longer opens with cv2.CAP_PROP_SETTINGS So basically what happened is that your source systems has another video capturing backend installed which is grabbed be |
2019-08-21 04:52:58 -0600 | commented question | Cross-compile for ARM Its somewhat out of my expertise and not really sure how to help you further... |
2019-08-21 04:49:45 -0600 | commented question | Python - Facial Recognition @holger, your comment makes absolutely no sense :D |
2019-08-21 04:48:38 -0600 | commented question | How to link with OpenCV as cmake subdirectory yep that seems correct, and indeed I guess you will need to build OpenCV first before your top project and not during th |
2019-08-21 04:47:54 -0600 | commented question | How to link with OpenCV as cmake subdirectory yep that seems correct |
2019-08-21 04:47:28 -0600 | commented question | OpenCV 4.1.0 not changing brightness Can you please use the correct flags instead of integer values, as described here? |
2019-08-21 04:46:58 -0600 | commented question | OpenCV 4.1.0 not changing brightness Can you please use the correct flags instead of integer values, as described (here)[https://docs.opencv.org/3.1.0/d8/dfe |
2019-08-21 04:40:06 -0600 | commented question | Using inpaint to restore old photo Well with mixing I mean alpha blending. Your border is so huge it has to inpaint pixels from far away. Also the borders |
2019-08-21 04:38:38 -0600 | commented question | Not able to identify the images correctly Ow your approach is sow flawed and for the following reasons haartraining is an old tool, it was replaced by traincasc |
2019-08-21 04:34:39 -0600 | commented question | error in command line That error is not due to OpenCV code but an output error of your realsense sensor, as described here ... |
2019-08-21 04:30:43 -0600 | commented answer | How to get areas of objects after watershed segmentation? You are retrieving the watershed regions as markers. These are basically points that have a specific ID. You can add all |
2019-08-20 08:58:03 -0600 | commented question | OpenCV face detection results in "debug assertion failed" Can you check if adding an explicit clone after the frame grabbing, so within the loop, solves the issue? Like a hard lo |
2019-08-20 08:54:18 -0600 | commented question | error in command line The only time I get such link errors is when I try static linking and order of libraries is incorrect. Is your library l |
2019-08-20 08:50:19 -0600 | commented question | How to generate a 3D image based on ChArUco calibration of two 2D images BTW, your approach is broken in the beginning with the matcher. It will never work since your calibration pattern covers |
2019-08-20 08:48:25 -0600 | commented question | How to get areas of objects after watershed segmentation? Have a look at the coin segmentation example using watershed. If you calculate the distance transforms, than threshold i |
2019-08-20 08:47:54 -0600 | commented question | How to get areas of objects after watershed segmentation? Have a look at the coin segmentation example using watershed. If you calculate the distance transforms, than threshold i |
2019-08-20 08:44:34 -0600 | commented question | Template matching with zero mean normalized cross-correlation _NORMED only means range normalized values in the case of the opencv cross correlation implementation. If I have a look |
2019-08-20 08:39:12 -0600 | commented question | The results of cv::eigen() are different on Win10 and Ubuntu. It seems, using the knowledge of this SO issue that data collected from source on 2 OS systems could have slight differe |
2019-08-20 08:28:07 -0600 | commented question | Different output in adding two images. If you really want people to dig into your issue, create a code sample to illustrate your problem and add it to the post |
2019-08-20 08:25:41 -0600 | commented question | Advice in learning OpenCV for android development There is a whole range of books on OpenCV for android and some of them target specific platforms like Android. I suggest |
2019-08-20 08:21:34 -0600 | commented answer | Dynamically choose best instruction set for hardware @Eduardo, that is what I mean, but the problem is that he is running on Windows, on which collecting that info is not as |