nice explanation! One suggestion - reduce text size to allow for a more compact view of your script. The column on the right looks microscopic by comparison
thanks man! you are the only guy explaining everything clearly and giving the source code. thanks! but I cant get the source code via your website , I wanted to download source though it is 20-30 lines of code :)
i really like your tutorial style. I'm curious if you've talked about working with much longer and higher resulution video. I've had trouble before using the naive opencv way of reading frames and processing at the same time. My program will crash. Imutils seems to be a popular approach but i dont see many yt videos about this
Thanks for the video, it's one of the best I've ever seen. There is a way to save these milestones in a database or in a csv file and then do a classification with a neural network. Do you have any videos you want? Thanks.
hi sir ur videos are too good these are helped me so much thanks for that.sir my point is every face is unique face and have unique facial coordinates points so y dont we have to store them points and used to recognise faces .if possible please doo a video sir.thank you in advance from sai kumar
I compiled and ran the same face recognition in the C++ environment, but when I recognized it under the camera, those points are always not stable enough. I tried cmake to compile the SSE4 command and published it as relaase, but I still see it clearly. The feature points jump, what is the reason for this? By the way, I am using a dell notebook.
Hi there... Currently, I'm working on my thesis and I want to ask something. Can I create a custom landmark predictor? If I can create my own, how to create it? because I'll be using it for my thesis. Then who create "shape_predictor_68_face_landmarks.dat" file? Is it from Python it self or who? Thanks btw 😁
hi! i ran the program and it works, but not as consistently as yours. the landmarks dont stay on my face when i move it around. I wanted to know if poor video quality is causing this? any suggestions for this? also, the detection does not work if i wrap a scarf around my head. Is there any way to detect the face in this condition? thanks a lot!!!
Awesome tutorial.... I am getting an error "RuntimeError: Unable to open shape_predictor_68_face_landmarks.dat" this python 2.7 and dlib installed ....where can i find that .dat file
hi, you need to download this file: github.com/AKSHAYUBHAT/TensorFace/blob/master/openface/models/dlib/shape_predictor_68_face_landmarks.dat and put it on the same folder where you have your python file
hey Pysource , were'nt you gonna post a tutorial on connecting opencv with Ip cam....did you abondoned that project or is it coming soon...btw i managed to connect the ip cam of android with opencv but only with the same network... i cant figure out how to do it when networks are different.
Yes, it's possible to measure the distance in pixels. you need to treat that as basic geometry and calculate the distance between two points: (x1, y1) and (x2, y2)
i followed your tutorial about how to install dlib, I downgraded my python version from 3.7.1 to 3.6.8 but it seems that it is not supported by opencv because i'm getting an error trying to install it. I would be thankful if you can suggest some solution
It's possible, though this currently is outdated as there are better methods out there, now in 2021. see this one among my latest videos th-cam.com/video/LF7Lgz4_lus/w-d-xo.html
Traceback (most recent call last): File "C:/Users/hitesh/PycharmProjects/DriverDrowiseness/faceDetectionSample2.py", line 12, in faces = detector(gray) TypeError: 'NoneType' object is not callable [ WARN:0] global C:\projects\opencv-python\opencv\modules\videoio\src\cap_msmf.cpp (674) SourceReaderCB::~SourceReaderCB terminating async callback hey pysource plz help me out
![ร้อยพ่อพันแม่ - WanMai [Official MV]](http://i.ytimg.com/vi/XY2pNYASEbg/mqdefault.jpg)
![PASULOL รามเกียรติ์ ตอนที่ 4 นนทกพบรัก [Ramakien Ep.4: Nonthok in love]](http://i.ytimg.com/vi/tMUQgmZzf_A/mqdefault.jpg)
It's the best. I mean I have never seen a person that shows these things on a simple tutorial. You're the best :)
So much better than doing this using Haar Cascades. It's virtually jitter free. I'm rewriting my code for dlib instead of haar now!
you explain it all in a very simple and easy way. Its very helpful.. Thank You
Became fan of you sir, your detailed explanation was awesome..want to learn more in computer vision from you..keep up the good work sir
It is very clean to understand and follow. Keep it a good job
Great tutorial!! Straightforward and easy to follow. Thanks!
Excelentes tus videos. He aprendido mucho con usted, muchas gracias y saludos desde Venezuela.
nice explanation! One suggestion - reduce text size to allow for a more compact view of your script. The column on the right looks microscopic by comparison
Thank you so much sir, i understand with your explain
Thank you very much! Very easy to understand and digestible!
awesome tutorial, been looking for this for a long time now.
thanks man! you are the only guy explaining everything clearly and giving the source code. thanks! but I cant get the source code via your website , I wanted to download source though it is 20-30 lines of code :)
Great video tutorial that I've ever seen.
Such a wonderful
The best tutorial. Thank you very much.
Great
thank you soooo much
keep up good work
You're awesome man! I learn it because of you.
Thanks! Simple and straightforward
Great job sir ...good explanation.
Thank you man, it was very helpful and I am looking forward to watch your other videos .............
Sooooo underrated sir thank you
Thank you very much sir. Your video helped me a lot.
i really like your tutorial style. I'm curious if you've talked about working with much longer and higher resulution video. I've had trouble before using the naive opencv way of reading frames and processing at the same time. My program will crash. Imutils seems to be a popular approach but i dont see many yt videos about this
Thanks for the video, it's one of the best I've ever seen. There is a way to save these milestones in a database or in a csv file and then do a classification with a neural network. Do you have any videos you want? Thanks.
Thank you for the tutorial.
Please do a face effect filter.
Thank you.
That was already in my plan :) I'm going to make it soon
It was really helpful! Thank You!
Great Tutorial
thanks
@@pysource-com Just Don't Stop on this opencv Tutorial it's soooo Usefull for me Thank's a lot
I am not able to get that file shape_predictor_68_face_landmarks.dat?? Please Help
superb. very simple and understandable. (Y)
Thanks so much man!!
can you make a tutorial of the face orientation?
many thanks.
Great tutorial , thank :
thank you soooooo much you are a life saver
Sir!!what content is present in shape_predictor_68_face_landmarks.dat file...
Good work
hi sir ur videos are too good these are helped me so much thanks for that.sir my point is every face is unique face and have unique facial
coordinates points so y dont we have to store them points and used to recognise faces .if possible please doo a video sir.thank you in advance from sai kumar
Super sir...
very nice video thank you
Hi
How can create bone and animation character in blender with opencv motion capture
thank you good sir
great content!
Teach you to use the detected value. About how to do it
I compiled and ran the same face recognition in the C++ environment, but when I recognized it under the camera, those points are always not stable enough. I tried cmake to compile the SSE4 command and published it as relaase, but I still see it clearly. The feature points jump, what is the reason for this? By the way, I am using a dell notebook.
Hi there... Currently, I'm working on my thesis and I want to ask something. Can I create a custom landmark predictor? If I can create my own, how to create it? because I'll be using it for my thesis. Then who create "shape_predictor_68_face_landmarks.dat" file? Is it from Python it self or who?
Thanks btw 😁
thanks man:)
which platform are you using to run this code
God blessed you
great job bro thx :)
Where is the link to download .dat file
how does it looks like, when you wearing a mask?
Thanks Bro
Thank You!!
This is great! I really appreciate your video. Is there a way to output this data to Blender, or to a file that Blender can read like fbx?
hi! i ran the program and it works, but not as consistently as yours. the landmarks dont stay on my face when i move it around. I wanted to know if poor video quality is causing this? any suggestions for this? also, the detection does not work if i wrap a scarf around my head. Is there any way to detect the face in this condition? thanks a lot!!!
please can i do this to get the points
left_top_pts, right_buttom_pts = face[0] , face[1] ?
Where can I get that .dat file?
After faces = detector(gray) there is no array in the output it's just showing a blank rectangle object
How we Apply this face landmark on real time capture?
where is the link for the shape predictor file
Hello good day .
where do you write all the commands?
Hi, I use pycharm
Tank youu
Awesome tutorial.... I am getting an error "RuntimeError: Unable to open shape_predictor_68_face_landmarks.dat" this python 2.7 and dlib installed ....where can i find that .dat file
hi, you need to download this file: github.com/AKSHAYUBHAT/TensorFace/blob/master/openface/models/dlib/shape_predictor_68_face_landmarks.dat
and put it on the same folder where you have your python file
hey Pysource , were'nt you gonna post a tutorial on connecting opencv with Ip cam....did you abondoned that project or is it coming soon...btw i managed to connect the ip cam of android with opencv but only with the same network... i cant figure out how to do it when networks are different.
Hi, I didn't abandon that! I will do that soon as well
@@pysource-com looking forward to it
how can i get dlib library in pycharm pleaze
please tell how to stop landmarks from shaking?
is it posibble to measure distance between 2 facial landmark? im learning it for my project :)
Yes, it's possible to measure the distance in pixels. you need to treat that as basic geometry and calculate the distance between two points: (x1, y1) and (x2, y2)
i followed your tutorial about how to install dlib, I downgraded my python version from 3.7.1 to 3.6.8 but it seems that it is not supported by opencv because i'm getting an error trying to install it. I would be thankful if you can suggest some solution
I have compiled python 3.7 version using cmake and vs 2019 . Search python packaging , sdist --wheel part .
getting an attribute error that dlib has no attribute 'get_frontal_face_detector'
I am getting error as dlib has no attribute get_frontal_face_detector()
how to get landmarks in numeric values using dlib
detector is not defined.. why ??? please help me guys
Is it possible to integrate this code with face recognition with neural networks?
It's possible, though this currently is outdated as there are better methods out there, now in 2021.
see this one among my latest videos th-cam.com/video/LF7Lgz4_lus/w-d-xo.html
This is really good
But how did you create that .dat file?
github.com/AKSHAYUBHAT/TensorFace/blob/master/openface/models/dlib/shape_predictor_68_face_landmarks.dat
Hi, how to hover one image on another image so that we can find coordinates of second image on first image and save that .
Use paint
How to deal with this error: "module 'dlib' has no attribute 'get_frontal_face_detector' " ?
have you find any solution for this ? I am also getting same error.
Traceback (most recent call last):
File "C:/Users/hitesh/PycharmProjects/DriverDrowiseness/faceDetectionSample2.py", line 12, in
faces = detector(gray)
TypeError: 'NoneType' object is not callable
[ WARN:0] global C:\projects\opencv-python\opencv\modules\videoio\src\cap_msmf.cpp (674) SourceReaderCB::~SourceReaderCB terminating async callback
hey pysource plz help me out
how to create those landmarks tho?
is it same as openface?
How i can create training model in this dlib???
dlib.net/ml.html#shape_predictor_trainer
In windows 10 why dlib module installed
how to install dlib?
check here th-cam.com/video/HqjcqpCNiZg/w-d-xo.html
play at 1.5 speed thanks me later
github.com/davisking/dlib-models/blob/master/shape_predictor_68_face_landmarks.dat.bz2 (link for the .dat file)
Windows ~> pip install dlib : Error
you can see here how to install dlib th-cam.com/video/HqjcqpCNiZg/w-d-xo.html
Just don't have python version greater than 3.6 and you'll be alright
为什么我运行后打开视频特别卡顿？