Ask Your Question

holger's profile - activity

2019-08-17 15:08:56 -0500 commented question kmeans in Android (Java)-

No - each cluster has a center - the so called centroid. Which is just the average of all points. The role of the labels

2019-08-16 09:25:35 -0500 commented question kmeans in Android (Java)-

Kmeans is a clustering algorithm. This means it will try to for groups of datapoints related to each other. Its does thi

2019-08-16 09:24:12 -0500 commented question kmeans in Android (Java)-

Kmeans is a clustering algorithm. This means it will try to for groups of datapoints related to each other. Its does thi

2019-08-15 15:00:13 -0500 commented answer Avoid retraining a model when executing a program?

overfit(use fine tuning / transfer learning from a face detection model) the face you want to detect - yeah maybe really

2019-08-15 14:59:04 -0500 commented answer Avoid retraining a model when executing a program?

overfit(use fine tuning / transfer learning from a face detection model) the face you want to detect - yeah maybe really

2019-08-15 14:58:06 -0500 commented answer Avoid retraining a model when executing a program?

overfit the face you want to detect - yeah maybe really absurd.

2019-08-15 13:55:52 -0500 commented answer Avoid retraining a model when executing a program?

Some time ago i had the idea of using overfitting to only detect one particular face- do you think this could make sense

2019-08-15 03:02:35 -0500 commented question What is wrong in this Opencv Face Recognition code?

detection: finding the position of your object (this is a face) recognition: identity the object (this is Pamela Anders

2019-08-15 03:01:43 -0500 commented question What is wrong in this Opencv Face Recognition code?

detection: finding the position of your object recognition: identity the object

2019-08-14 02:42:29 -0500 commented answer I am getting bug while performing open cv based dnn face detection

@Shiva_2410 From what i understand from berak "give a proof it worked": Make sure the mat is loaded correctly and print

2019-08-12 11:00:14 -0500 commented answer Cannot use Tesseract with OpenCV 4.1.1

And its not working... I am giving you a working solution - at least its verified for my setup - your turn now. Actuall

2019-08-12 10:47:22 -0500 commented answer Cannot use Tesseract with OpenCV 4.1.1

And its not working... I am giving you a working solution - at least its verified for my setup - your turn now. Actuall

2019-08-12 09:24:05 -0500 commented answer Cannot use Tesseract with OpenCV 4.1.1

pom.xml - maven dependency file - or how do you manage your dependencies?

2019-08-12 08:29:30 -0500 answered a question Cannot use Tesseract with OpenCV 4.1.1

Add this two lines to your pom - this works for me (window 10 64 bit) <dependency> <groupId>org

2019-08-12 08:11:42 -0500 commented answer OpenCV 4.1 + Python 3 + CUDA

Yes for a single image with a simple operation - using a gpu is maybe even slower than on cpu. you need to transfer the

2019-08-11 11:22:14 -0500 answered a question OpenCV 4.1 + Python 3 + CUDA

Look you touching a "burned" area in opencv now - and you should just invest your time into better things imho: Let me t

2019-08-11 10:55:18 -0500 commented question how to solve problem of low fps in OpenCV

Use a model based on a convolutionary neuronal network - just google it and pick a framework with gpu support (they all

2019-08-09 15:38:31 -0500 commented question Compare segmentation result to ground truth

Well you need some metric like this https://www.pyimagesearch.com/2016/11/07/intersection-over-union-iou-for-object-dete

2019-08-09 15:33:22 -0500 commented answer OpenCV doesn't work with two python versions

Look pls- just stay calm and use your head. What i t tried to tell you is to just reinstall opencv for the version where

2019-08-09 14:26:57 -0500 answered a question OpenCV doesn't work with two python versions

Haha :-) Well that's the python madness of installing dependencies per python installation - globally-. Just reinstall

2019-08-07 15:17:02 -0500 commented question how to solve problem of low fps in OpenCV

Use a cnn model with a nvidea gpu.

2019-08-07 15:14:30 -0500 commented question Image recognition with a single reference picture

berak gave the right keyword - for your problem you can check for facenet. I even have a facenet micrososervice in pytho

2019-08-06 18:26:43 -0500 answered a question Region of interest extraction

Don't know - maybe use contours or a object detector to get the shape or bbox for your object and then just crop.

2019-08-02 14:58:04 -0500 answered a question how to classify a photoset

I recommend coursera - i am currently doing this course https://www.coursera.org/learn/machine-learning/home/welcome But

2019-08-02 14:44:10 -0500 commented question Exception in thread "main" java.lang.NoClassDefFoundError: org/opencv/core/Core

You are welcome. Good that the opencv user forum is so friendly ^^

2019-07-29 03:40:07 -0500 commented question Exception in thread "main" java.lang.NoClassDefFoundError: org/opencv/core/Core

Hmm this is a ClassNotFoundException - not a linkage error or anything, so your code from above wil not change anything.

2019-07-26 10:46:12 -0500 answered a question OpenCL Performance (or lack of)

My own experience(highly subjective): Opencv has the fastest algorithm for running on cpu(high optimized) - if it does

2019-07-26 06:35:58 -0500 received badge  Self-Learner (source)
2019-07-25 10:52:24 -0500 marked best answer newbie question: working with higher dimensional mat

Hello,

I found some c++ / python code for evaluating a text detection model -

c++ https://github.com/opencv/opencv/blob...

const float* scoresData = scores.ptr<float>(0, 0, y);

python https://github.com/opencv/opencv/blob...

scoresData = scores[0][0][y]

These two snippets do the same in different programming programming language - its get back an array with probabilities. I am trying to port this call to java - but i am failing hard here.

Can you give me a small hint how to port this line in question to java? Thank you very much + Greetings, Holger

2019-07-25 10:51:21 -0500 answered a question newbie question: working with higher dimensional mat

Hmm a bit searching found me this https://stackoverflow.com/questions/53402064/opencv-east-text-detector-implementation

2019-07-25 10:28:40 -0500 asked a question newbie question: working with higher dimensional mat

newbie question: working with higher dimensional mat Hello, I found some c++ / python code for evaluating a text detect

2019-02-05 10:47:45 -0500 commented question Can someone explain the output of forward() in DNN module?

For classification - the output is usually a matrix with the probabilities for each class. For regression its usually a

2019-01-27 07:22:39 -0500 commented answer read formated matrix from text file

well if i rember right - in python its easy to serialize / read numpy arrays.You have this functionality because of nump

2019-01-23 12:51:46 -0500 commented question How align face?

thank you philli

2019-01-22 13:09:35 -0500 commented question How align face?

Yes this is the opencv version of the problem. Thank you for link,

2019-01-22 10:44:09 -0500 commented question want to detect rectangle in a image using opencv-python?

you have width and heigh of rect - so it should be easy to decide which one is bigger - essentially what berak posted wi

2019-01-22 10:43:48 -0500 commented question want to detect rectangle in a image using opencv-python?

you have width and heigh of rect - so it should be easy to decide which one is bigger - essentially what berak posted wi

2019-01-22 10:38:36 -0500 commented question How align face?

I found some time ago some code on pyimagesearch - that adrian guy gave a solution for this. And if i remember right - i

2019-01-22 10:10:09 -0500 commented question Can dnn model made under Python be used under JAVA

Yes this should be possible. Open cv can read and evaluate several models. The way the models were created (python, c++

2019-01-18 11:19:50 -0500 commented answer SVM on CUDA

@berak Ich bin aus Brandenburg und wohne in Berlin. Freundliche Grüße.

2019-01-18 10:58:40 -0500 commented answer SVM on CUDA

Berak kommst du aus Deutschland?

2019-01-18 04:50:40 -0500 commented question Image Classification

Ok got it - so its a certain model configuration with weights trained on imagenet. I know imagenet is biiig so i guess m

2019-01-16 16:37:17 -0500 commented answer How to calculate the head and body angle with respect to the camera?

You guys are awsome.

2019-01-16 16:31:15 -0500 commented question Image Classification

@StevenPuttemans What is a darknet48 network? Is it a yolov3 config stripped down to 48 layers and the weight were cre

2019-01-16 16:23:11 -0500 commented question Image Classification

@StevenPuttemans What is a darknet48 network? Is it a yolov3 config stripped down to 48 layers and the weight were cre

2018-12-24 22:03:03 -0500 received badge  Teacher (source)
2018-11-30 08:11:05 -0500 commented answer Keras DenseNet121 breaks on opencv dnn

Just use the native dnn?

2018-11-30 08:02:05 -0500 commented question Opencv load tensorflow Style Transfer Model Failed

Well somone has the same problem like - external model with opencv fails. This time its yolo and not tensorflow but sam