Make Your Own Haar Cascade Object Detector On Windows | QUICK AND VERY EASY

[@adityapaithon6499]

Realtime applications of Haar Cascades for Object Detection with Python Code : kzbin.info/www/bejne/hKKreJKtjNehhpY Subscribe! For more videos on Python❤

[@ZirbertEruj]

Loud and clear, straight to the point. Subscribed!

[@IKMCDANIEL]

Fast, efficient, clear. Excellent video!

[@adityapaithon6499]

Thank you for the feedback Daniel :)

[@nalininadupalli364]

This is by far the best explanation of Face detection based Viola-Jones method I've come across. Excellent, to the point and simply best of them all. Thank you so much for your video!

[@adityapaithon6499]

Welcome, glad to help

[@NS-zi1xv]

Its always the indians who save the day lol nice video dude

[@MrLLGO]

Thanks Bro!!!! Saved my Life!! Greetings from Germany!!!

[@alonewolf7682]

thanks man saved me from college mini project

[@krishnapriyanair1063]

hey I have used the same for detecting a scratch in the object but after running the test I didnot get the classifier file

[@mansudshaik7964]

How to improve accuracy of detection

[@anubhavpal5782]

what's the maximum size limit because keeping width and height above 70 crashes or hangs my pc ? please let me know

[@kaanankalyan7779]

Helped a lot, thanks

[@sampc9542]

U have explained very well bro keep it up

[@46sauravdubey31]

Excellent video

[@jordanwilliams851]

I had an error that said insufficient memory and to fix it on the input tab i changed the negative image count from the default 1000 to the exact number of negative images I have in my "n" folder hope this helps others :)

[@adamt7367]

I had this issue too but I accdetnly downloaded the 32bit version, make sure you have the right version all!

[@thushiyanthk1668]

Thanks bro

[@ritikchaurasiya5588]

Bro I got an error pls solve it ....there is an error like .....Open CV error:Bad Argument(cannot get new positive sample the most probable reason is insufficient count of positive samples in a vec- file.

[@robodev6033]

Very useful video thank you so much .

[@akbarwijaya7781]

thanks bro, you saved my life :'

[@emanueleaets9316]

To have better performance, do the positive image have to be of good quality or with few details over the rilevant details. For example Is It enoug Only the frame with eyes, nose, mouth, eyebrows, and leave out the hair and the landscape back the face ? Thank you for everything

[@sahanlahiru7193]

thank you❤

[@diseaseddates4242]

amazing video ty!

[@p1llai]

At the end it says Open Cv memory error something please help It’s urgent for my project.

[@gabrielrenno104]

Did you solve this? same here

[@wanfadhliazmin]

look great and promising for beginner like me. do you have blog for this tutorial?

[@kushalbhavsar3454]

Thanks dude!!!you saved my lots of time:))))

[@lac19951]

Hey Aditya, thanks for this technique. What about detecting false positives?

[@hindiwebseries4321]

Sir in folder only positive and negative images should be there or any other file is also required

[@sumanthsrivatsa3249]

Nice video Aditya approximately how much time it takes to train totally

[@adityapaithon6499]

Depends on how many images youve got and system specs

[@sumanthsrivatsa3249]

@@adityapaithon6499 Aditya the GUI which you have specified I have tried but it trained the data only once I am not sure whether the problem is with the application or my laptop

[@Prachi_Mayank]

Can u plz share your last video's link which you are talking in that

[@adityapaithon6499]

kzbin.info/www/bejne/hKKreJKtjNehhpY

[@bunnyvlogs7647]

Ur really great brother. ..God bless u

[@adityapaithon6499]

Thanks Sai

[@bunnyvlogs7647]

We suggest u similar explaination on YOLO v4 framework from scratch....if possible...

[@rasanjalirathnayake8873]

OpenCV Error: Bad argument (Can not get new positive sample. The most possible reason is insufficient count of samples in given vec-file. ) in CvCascadeImageReader::PosReader::get, file D:\cv\opencv_3.2.0\sources_withTextModule\apps\traincascade\imagestorage.cpp, line 158 PLEASE HELP TO RESOLVE THIS

[@mikaelvi3960]

Thanks bro for this amazing tutorial :)) I thought the negative images are the images that are gray scaled that won't contain object being detected and the positive are the images being detected. can please explain? or it works the same?

[@chageruvinaykumara4949]

Am not getting proper output...plz can u explain y ??

[@tejatechviews]

Please provide the data set which is not opening the website for face data

[@ishan_shrestha]

Awesome Video. Simplest out there. During training what type of images to you suggest? One with plain background or one with various stuffs in the background.

[@MH-zq4br]

Mine doesnt generate any xml file after training

[@powerpower1334]

Train dataset for temp stage can not be filled. Branch training terminated. please help me solve this error

[@kanchanaramar]

Thank you. Just what I was looking for!

[@krishnasoni5883]

Absolutely fantastic , thanks for this

[@amrithaasivakumar5200]

Instead of getting a single box outlining the image during test, i get several small sized boxes all over the picture. Pls do help.

[@jaypanchal3875]

amazing explanation bro...... thank you..... but an error found as insufficient memory to allocat 1468024264 bytes , when my disc is 23 gb free?? can you solve this??

[@sundarnatarajan7431]

Excellent dude.You actually saved lot of time.

[@adityapaithon6499]

Thanks Sundar

[@sundarnatarajan7431]

Any idea how amin did the GUI ?

[@emanueleaets9316]

Hi Sir thank you a lot for your job, this is the best video tutoria i have found please can you help me? I have one question, the size of positive images have to be of a specific value? When you select the height and width' size, does it refers to the rectangle that will contain the detected face? Sorry if my English is not much good

[@mnjinderpawar4029]

Awesome video

[@orvillethomas4464]

you the best bro !!

[@sala7311]

Great video, Well explained Bro!

[@adityapaithon6499]

Thanks :)

[@radhakrishnanganapathy899]

n and p images trained but xml file will not created . OpenCV Error: Insufficient memory (Failed to allocate 2147493236 bytes) in cv::OutOfMemoryError, file D:\cv\opencv_3.2.0\sources_withTextModule\modules\core\src\alloc.cpp, line 52. what to do? what format and resulation want to choose ?

[@rishipareek4803]

I want to make haarcascade classifier for identify cow 🐄

[@adityapaithon6499]

Why?

[@samueleboh8965]

I really appreciate. Thanks man.

[@HiepNguyen-gq5lp]

Do the images in the 'n' and 'p' folders need to be the same size? thanks so much!

[@eyuelworku5744]

Have you found that out? I was wondering the same thing

[@anubhavpal5782]

@@eyuelworku5744 what's the maximum size limit because keeping width and height above 70 crashes or hangs my pc ? please let me know

[@emanueleaets9316]

Hi sir, Thanks for your work. Can you help me please ? I'm trying to apply your guide to detect face in picture and via webcam, but lots of other object are detected How can i reduce the noise ? I used 300 positive images, sometimes i used also 2k positive images

[@maverickgaming4659]

I also have the same problem, have you found a solution?

[@v_hackkali8634]

What i can do? OpenCV Error: Unknown error code -9 () in cvSetImageROI, file D:\cv\opencv_3.2.0\sources_withTextModule\modules\core\src\array.cpp, line 3021, OpenCV Error: Bad argument (Can not get new positive sample. Seems that vec-file has incorrect structure. ) in CvCascadeImageReader::PosReader::get, file D:\cv\opencv_3.2.0\sources_withTextModule\apps\traincascade\imagestorage.cpp, line 161.

[@ilkeraykut7064]

I wanna do this for electronic devices lke mosfet or resistor.I tried this tutorial but it was not efficient.What am I supposed to do?Thanks.

[@adityapaithon6499]

Look into something called "Convolution Neural Networks"

[@barathkumarg9197]

sir i have created a cascade for animal detection the xml file created successfully but the detection is not working properly how can i do it

[@bryanltobing]

How long will it take until the training is over?

[@sanishbabu2435]

I just tried to make my own haar cascade file of bags as u told but it is not detecting bags and detecting random objects what should I do (real-time bag detection from laptop cam )

[@rabiyatahir5530]

very very very helpful video

[@milanjayan1384]

Good presentation with lot of information.

[@MachineWala1]

only cascade file is not generated what can i do? help me

[@adityapaithon6499]

Attach logs.

[@hermaDeWiDy]

excuse me, what should i do for : OpenCV Error: Insufficient memory (Failed to allocate 2147492516 bytes) in cv::OutOfMemoryError, file D:\cv\opencv_3.2.0\sources_withTextModule\modules\core\src\alloc.cpp, line 52

[@TejasPhase]

Hello Aditya Sir, What is "sample width" & "sample height" parameters? If the image is 640 x 480 pixels then what will be those width & height parameters?

[@priyankag4442]

Can you please suggest us a classifier for detecting cracks ?

[@adityapaithon6499]

Cracks? What do you mean lol

[@devopsengineers8254]

Amazing video brother

[@adityapaithon6499]

Thanks Shivam!

[@abderrahmenmelliti9627]

hey in folder classifier i find only log i can't find cascade file what's the problem ?

[@adityapaithon6499]

I think the path provided is not accurate

[@manjunath.c2944]

same error how to fix it???

[@AdityaKumar-zf4cf]

Thnx for Video.I have few questions: So here we are using Voila Jones as it uses Haar Cascade? So No coding required to make your own Haar Cascade just that its application can be simply used as other OpenCV Cascades?

[@adityapaithon6499]

Correct.

[@Abhishek10p]

Can u plz send link of CK database...

[@adityapaithon6499]

www.kaggle.com/c/visum-facial-expression-analysis

[@Abhishek10p]

@@adityapaithon6499 Thx

[@sanju3937]

Awesome one buddy

[@adityapaithon6499]

Thanks

[@kramesh5365]

Bro we are really happy for u r valuable time which was spent for us. Thanks a lot I need detailed information about the program which is used in this video about each and every function how it works. please help me if its possible.

[@adityapaithon6499]

Checkout my other video "Detect Cars in video" in the pinned comment

[@DrSirmacek]

Such a nice video!

[@aamirahmadansari2721]

Hello bro can we connect, actually I am unable to train my haar for palm detection, it says bad argument can't have a new positive values and I am using 1000 negative and 2800 positive images

[@firstviewoftrading]

Nice

[@aaditpatel3032]

Sir can you help ?I want to detect live faces but this model is unable to detect many faces even if i trained the model

[@mikaelvi3960]

Hey Bro thank you for the educational video tutorial, why do you have to grayscale your positive images?

[@zakariaabderrahmanesadelao3048]

subbed thank you for the awesome tutorial

[@bikramkarmakar6482]

How to decide Resolution of images(sample width and height)? please help me.

[@electroware9390]

Thanks a lot for this video

[@sangvo2620]

I have one question about Cascade Trainer GUI - How can I properly stop and resume the training? Despite loading the same setting from .ini file and press start, it starts from scratch.

[@lezginkaya8138]

Thanks for this useful video

[@AbhishekSharma-hy4nl]

Not working..it's showing "opencv error : insufficient memory"... Someone please help me out..

[@oledgaming8011]

How much time does it take to complete 100%?

[@aiemporium8367]

Is this process applied for positive color images?

[@tarundeshmukh8062]

Anyone have Harcasscade xml file for helmet detection?

[@adarshsrivastava7168]

Sir, How can I make haar cascade file for the personality prediction

[@irfanhassan2103]

OpenCV Error: Bad argument (Can not get new positive sample. The most possible reason is insufficient count of samples in given vec-file. ) in CvCascadeImageReader::PosReader::get, file D:\cv\opencv_3.2.0\sources_withTextModule\apps\traincascade\imagestorage.cpp, line 158 this error is coming and i am stuck there

[@amirsohail963]

data set CK dataset LINK not working can change UPDATE link if possible.

[@balajii5899]

Thank u ,u r awsm😍

[@adityapaithon6499]

:) ❤

[@dylansanderson3386]

hello. Does anybody know, do the positive images have to be in a folder named exactly "p" and negative in folder "n"?

[@nicemachado5976]

Nice tutorial! What if i want to classify different types of objects like apple, watermelon, pineapple etc? Does this program support more than one object at once?

[@muhammadosama3358]

You need a neural network for that.

[@mohammedazzan7529]

@@muhammadosama3358 how do we do that? is it easy? or difficult?

[@muhammadosama3358]

@@mohammedazzan7529 very easy, there are readily available frameworks for that. for starters you can use sklearn (python library)

[@mohammedazzan7529]

@@muhammadosama3358 omg thanks a lot for the reply man I was really scared when I picked up this project 😭 I just want to make a project to detect different fruits that's it do you think it's possible?

[@muhammadosama3358]

@@mohammedazzan7529 yes, you can even get the dataset from kaggle, UCI Irvine repository, tensorflow hub etc

[@x1WARLOCK1x]

When I use more than 332 positive images i get an Error, that opencv couldnt fetch a new p-image because it isnt specified in the vec-file. Could you help or have an idea?

[@ashutoshtiwari3785]

What screen recorder are you using?

[@sanjayjha35]

Hi Aditya, does creating or training these kind of cascade XML files for training purpose comes under data science(ML)?

[@kushagrababbar1395]

I am using 7 pos and 200 neg images and trying 15 stages but it stops at 2nd stage with this error . -----> Required leaf false alarm rate achieved. Branch training terminated. Please help me in this problem sir .

[@RnFChannelJr]

i get some error when i train more than 2000 images , what should i do ??

[@AakashGhodke]

i run it twice it create only classifier folder log etc but not the .xml file

[@adityapaithon6499]

How much time did it take to train? Also Open the log and search for any errors

[@utkarshtripathi6123]

@@adityapaithon6499 same error log file does not show any error, for one dataset it took 2.5 hours and for one smaller data set it took 1 hour.

[@utkarshtripathi6123]

@Zendevil set the height and width as shown in video, it solved my problem

[@manjunath.c2944]

same problem how to fix this error ??

[@mohanasarvanipalla]

wat about color images i want to implement it using color images

[@adityapaithon6499]

Use Colour Segmentation Tutorial : kzbin.info/www/bejne/Y3i7m2iHbtasY5I

[@rupeshghimire3365]

I have 4000 negative images. Do I need to change the negative image count to 4000 ?? Can I train the model with different image formats at once ??

[@deeraj3069]

Thank you.

[@ankursoni8060]

I created HAAR to detect mobile, I took 50 positive image and 50 negative image and after generation XML when I use that XML then I'm facing this exception opencv: the node does not represent a user object (unknown type?) Can you please tell me what is the error and how to resolve it

[@krishnamitabanerjee4351]

I'm using this way to implement pipeline leakage detection using 80 positive and 160 negative images,but not getting a decent output, how to overcome this?

[@abhishekray3460]

Can you pls share Train and test code as shown in video?

[@adityapaithon6499]

Yes it's in my other video on channel

[@circuitdesigner7489]

is it necessary to download grayscale images

[@souradipbhadra1406]

is it neccesery to convert picture color into grayscale?