BIG NEWS FAM I used the stuff I learned from you in a school interview and I am now part of the robotics team! :D Thanks for teaching me everything I know!
Excellent Nick.. You continually break down complex topics into layman's terms. Becoming a patreon of your content is one of the best instructional investments I've ever made.
Hello Nick, this playlist is just awesome!! I was wondering if you could please make a similar paper to code series for the FaceNet paper with training on Triplet Loss... in specific to implement triplet loss on your own dataset of images and to learn how to make the triplets (hard and semi-hard triplets) and how to implement and evaluate different data mining strategies (offline vs online) ... I'm crossing my fingers and hoping to see something on this topic from you
Hey Nicholas - tensorflow and tensorflow-gpu seem to no longer be compatible due to version changes... Maybe you could post an update in terms of what libraries and versions should be used (or is the entire project deprecated as a result?). Thank you!
Collecting grpcio Using cached grpcio-1.60.1-cp38-cp38-win_amd64.whl (3.7 MB) Installing collected packages: grpcio Attempting uninstall: grpcio Found existing installation: grpcio 1.32.0 Uninstalling grpcio-1.32.0: Successfully uninstalled grpcio-1.32.0 Successfully installed grpcio-1.60.1 ERROR: tensorflow 2.4.1 has requirement grpcio~=1.32.0, but you'll have grpcio 1.60.1 which is incompatible. ERROR: tensorflow-gpu 2.4.1 has requirement grpcio~=1.32.0, but you'll have grpcio 1.60.1 which is incompatible. Any ideas?
I study at the university the computer department, but I do not have a computer, I attend electronic lectures and exams on my mobile. I wish I had a computer, I work to buy it, but it is expensive 😞😞😞
@@NicholasRenotte Yes Nick. I am working on a project where I need to compare different industrial items. I have used an object detector when the items was less, now it is going to be more. So I am exploring this pipeline.
This playlist is great but there is a problem at the very begininng you start installing the dependencies but you haven't tell how to setup the conda or jupiter environment, if a person is starting from scratch it would be difficult for him to grab the basics
Hi Nicholas~ Happy new year and awesome tutorial! Revisiting one of your past videos and manage to create this Face Application over the holiday period Just wondering, how do we deploy the custom TF model into production? The h5 model is huge (not able to upload to Github) after data augmentation.
When I run the first Model(...) row I get the error "NameError: name 'inputimage' is not defined". I don't see where you define this previous to that row. Anyone knows why this error pops up?
if there are group of people on the image , i need to recognition particular face on that image that i trained in the model with there name and the rectangular line on the face , with accurate 90.1. can you put video for that using opencv ....please
how to solve this ERROR: Could not find a version that satisfies the requirement opencv (from versions: none) ERROR: No matching distribution found for opencv
Maybe an outdated version or no version at all of opencv try a "pip install opencv-python" or if you have anaconda "conda install -c conda-forge opencv"
What @SpicyCactus said is best, could also try using a different version of Python 3.7.3 is the version that I'm currently using. It's sufficiently advanced to pick up most new libs but no so far ahead that nothing works.
Hi renotee, Can we detect and recognize in Multiple people. Mean to say 100 of people we recognize 30 people , recognize true and 70 not by this approach if not so which Approach i have to use? Thanks
Lert say a label creates this app to verify the staff working in the offices. Logically for every member the model must be re trained. SO let say if there are 50 employees , the model must be trained 50 times. is that correct???
Hello Nicholas I completed the whole tutorial until part 7, everything worked perfectly fine, but when i am doing the inference, it’s verifying negative images are true as well, since there’s no error i m unable to trace it back to where might the problem. Is it because of the data quality?? My webcam is from the 90s, so not that clarity in my anchor images.
i have a problem in the part of GPU Growth , i think because of the tensorFlow version and tried another modern versions but the same error that it doesnt recognize any GPUs on the actual machine
I've watched your "Getting Started with Python Deep Learning for Beginners" video and then started doing this project, but when I want to execute and use the facial recognition app it gives me this error: "ERROR: Could not find a version that satisfies the requirement tensorflow==2.4.1 (from versions: 2.12.0rc0, 2.12.0rc1, 2.12.0, 2.12.1, 2.13.0rc0, 2.13.0rc1, 2.13.0rc2, 2.13.0, 2.13.1, 2.14.0rc0, 2.14.0rc1, 2.14.0, 2.14.1, 2.15.0rc0, 2.15.0rc1, 2.15.0, 2.15.1, 2.16.0rc0, 2.16.1) ERROR: No matching distribution found for tensorflow==2.4.1"
@@swarupsahu8560 I gave up on the project, then started working on google colab and chatgpt that way it was way easier to solve compability problems etc.
I have the following error code on colabs and I wonder if you can tell me what is wrong, please: !pip install tensorflow-gpu Collecting tensorflow-gpu Using cached tensorflow-gpu-2.12.0.tar.gz (2.6 kB) error: subprocess-exited-with-error
× python setup.py egg_info did not run successfully. │ exit code: 1 ╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip. Preparing metadata (setup.py) ... error error: metadata-generation-failed × Encountered error while generating package metadata. ╰─> See above for output. note: This is an issue with the package mentioned above, not pip. hint: See above for details.
In the further parts of this series, we take up 3000 image pairs for training. But won't it overfit a siamese network based model? Also, we were supposed to make a one-shot model which would take only one image (or probably very few images) for training the model and then for verification too. But here we take as much as 3,000 image pairs for training and 50 images for final verification. Kindly clarify a bit over these matters.
Hey, are you following the tutorials end-to-end? If yes, could you please share what should I do when if I'm coding for creating the directories, it is showing that they already exist. I tried experimenting with system environment variables, but unable to figure out the actual discrepancy. Also, the images and everything that my webcam is capturing is getting stored but it's not clear in which path!! Kindly help!
@@manasisingh294 Ig you can change all the paths and directories in the ipynb files (if you just want to run the project) and if you follow all the tutorials step by step, then you can create directories as you want. But if you are facing some problems, I would suggest follow what Nicholas is doing.
I had an erro while installing the first dependency (!pip install tensorflow==2.4.1 tensorflow-gpu==2.4.1 opencv-python matplotlib) i've already checked the requirements to installed and i have all so if you had the same problem i currently looking for a solution.
Hey, I love your videos and was trying to follow along, but for some reason, tensorflow won't detect my GPU despite reinstalling both the original version and the gpu version multiple times. do you have any fix for this?
Hello Nicholas Renotte, From your this video (th-cam.com/video/06TE_U21FK4/w-d-xo.html&lc=UgybAvzcbfF4RvWJTj54AaABAg.9R3kZbbuxsV9R8MRuJWJzm&ab_channel=NicholasRenotte) I have a question related to angle measurement, I want to get the angle for bowling action analysis... For that I need to calculate an angle between Shoulder, elbow and hand wrist... Form this video the angle is continuously measured from the camera but I want to calculate angle only when user perform a bowling action. For the ICC rule, the angle is measured such that first we calculate angle once when user arm is parallel to ground (angle1) and when he throughs the ball (angle2) , after that I need to subtract the angle2 from angle1.... Please guide Thanks
Could apply a pose detection model over the video first to detect when a player is in bowling then apply the angle estimate. Take a look at the body language decoder vid for how to do the first bit.
BIG NEWS FAM
I used the stuff I learned from you in a school interview and I am now part of the robotics team! :D
Thanks for teaching me everything I know!
GOTTTTDAMMMNNN, YESSS!!! OH man, I'm so happy for you. Congratulations!!!!
nice i love this to
The most well structured channel on youtube in terms of quality of content and smoothness of learning, thank you so much!
Looking forward to upcoming videos in this series
Ay man I was waiting for a video like this I appreciate it. Trying to work on a facial recognition NLP bot so looking forward to this
Sweet! Awesome stuff man!
@@NicholasRenotte Thank You, your videos are in awesome format very helpful!
The way you explain everything is amazing. ..
Excellent work. ..very easily explained
First comment I can remember ever making… never stop, this is the best format ever. Thank you so much for demo on everything step by step.
Looking forward to this series too nick.
YESS! I think building models from papers is such an awesome task!
I been waiting for a face recognition tutorial
Thank you
can you please make a video on face recognition using a faceNet model
Excellent Nick.. You continually break down complex topics into layman's terms. Becoming a patreon of your content is one of the best instructional investments I've ever made.
Thanks sooo much @Shane, love having you here!!
I hope next episode of this series comes soon!
This Sunday!
@@NicholasRenotte uohgbibggggggggggvvvvvvvttttttttttttttttttttttooooooooooooooooooooooooooooooooooooooooooo
Just the day I needed it! Excellent one! Waiting for next parts patiently.
Thank you for the great work and effort you are doing!
Great tutorial as always. Can't wait to see the next video. 😎😎
Coming soon @Alexandre!!
Yo Bro waiting for the other half. To be honest bro thanks for the paper links ur sharing thy r very useful
I have GPU in my laptop but it is showing 0 when I enter gpus command. What should I do?
activate cuda or 0 es "the GPU 0", it is one
Wow, awesome!!! Very exciting. TQVM!!
Hello Nick, this playlist is just awesome!! I was wondering if you could please make a similar paper to code series for the FaceNet paper with training on Triplet Loss... in specific to implement triplet loss on your own dataset of images and to learn how to make the triplets (hard and semi-hard triplets) and how to implement and evaluate different data mining strategies (offline vs online) ... I'm crossing my fingers and hoping to see something on this topic from you
best teacher...nice really algo to code is nice thing and 💡
Thanks so much @Saurabh!
great work dude, you're our savior
Hey Nicholas - tensorflow and tensorflow-gpu seem to no longer be compatible due to version changes... Maybe you could post an update in terms of what libraries and versions should be used (or is the entire project deprecated as a result?). Thank you!
hello, did you found any workaround? im stuck at same problem
Collecting grpcio
Using cached grpcio-1.60.1-cp38-cp38-win_amd64.whl (3.7 MB)
Installing collected packages: grpcio
Attempting uninstall: grpcio
Found existing installation: grpcio 1.32.0
Uninstalling grpcio-1.32.0:
Successfully uninstalled grpcio-1.32.0
Successfully installed grpcio-1.60.1
ERROR: tensorflow 2.4.1 has requirement grpcio~=1.32.0, but you'll have grpcio 1.60.1 which is incompatible.
ERROR: tensorflow-gpu 2.4.1 has requirement grpcio~=1.32.0, but you'll have grpcio 1.60.1 which is incompatible.
Any ideas?
please, Can you make videos about vehicle detection and tracking and explain new techniques in this field
I study at the university the computer department, but I do not have a computer, I attend electronic lectures and exams on my mobile. I wish I had a computer, I work to buy it, but it is expensive 😞😞😞
You are really a great teacher
Awesome Nick. Your work is really amazing.
I am about to ask about the same series and same paper
Niceee! Building anything cool?
@@NicholasRenotte Yes Nick. I am working on a project where I need to compare different industrial items. I have used an object detector when the items was less, now it is going to be more. So I am exploring this pipeline.
@@DrShaikAhmad niceee!
Thank you so much for share your knowledge man, u r the best.
This playlist is great but there is a problem at the very begininng you start installing the dependencies but you haven't tell how to setup the conda or jupiter environment, if a person is starting from scratch it would be difficult for him to grab the basics
Yes i simply tried to use to jupyter online notebook and faced difficulty installing libraries. How do i overcome that?
can you please make a video for beginners how to read research paper and implement them.
"Paper to code" => more please!
Hi Nicholas~
Happy new year and awesome tutorial!
Revisiting one of your past videos and manage to create this Face Application over the holiday period
Just wondering, how do we deploy the custom TF model into production?
The h5 model is huge (not able to upload to Github) after data augmentation.
When I run the first Model(...) row I get the error "NameError: name 'inputimage' is not defined". I don't see where you define this previous to that row. Anyone knows why this error pops up?
Is the variable defined inputimage or input_image?
love this series
that's good sir.
please Make a learning video about Emotion Detection or Emotion Recognition.
thank you.
Oooooh yeah, will add that one to the list! Have you checked this out? th-cam.com/video/We1uB79Ci-w/w-d-xo.html
if there are group of people on the image , i need to recognition particular face on that image that i trained in the model with there name and the rectangular line on the face , with accurate 90.1. can you put video for that using opencv ....please
great work here
can you make a video for fingerprint recognition or matching using siamese
Can you please tell me which language did you use to develop
how to solve this ERROR: Could not find a version that satisfies the requirement opencv (from versions: none)
ERROR: No matching distribution found for opencv
Maybe an outdated version or no version at all of opencv try a "pip install opencv-python" or if you have anaconda "conda install -c conda-forge opencv"
What @SpicyCactus said is best, could also try using a different version of Python 3.7.3 is the version that I'm currently using. It's sufficiently advanced to pick up most new libs but no so far ahead that nothing works.
@@NicholasRenotte now it is showing ERROR: Could not find a version that satisfies the requirement tenserflow-gpu==2.4.1 (from versions: none)
@@SpicyCactus7
now it is showing ERROR: Could not find a version that satisfies the requirement tenserflow-gpu==2.4.1 (from versions: none)
@@monstergamerz2343 pip install tensorflow-gpu==2.4.1
Hi renotee, Can we detect and recognize in Multiple people.
Mean to say 100 of people we recognize 30 people , recognize true and 70 not by this approach if not so which Approach i have to use?
Thanks
So, if I have a macbook air m1, I can skip the part of setting GPU growth?
Lert say a label creates this app to verify the staff working in the offices.
Logically for every member the model must be re trained. SO let say if there are 50 employees , the model must be trained 50 times.
is that correct???
Hello Nicholas
I completed the whole tutorial until part 7, everything worked perfectly fine, but when i am doing the inference, it’s verifying negative images are true as well, since there’s no error i m unable to trace it back to where might the problem.
Is it because of the data quality??
My webcam is from the 90s, so not that clarity in my anchor images.
dear can i have the source code??
yes we want to see this
sir I have a gpu in my laptop (rtx 3070 ti to be exact ), but somehow tensorflow is recognising my gpu
how can I fix this and use my laptop gpu
Hi bro can we connect?
please share the dataset also its also very helpful
Is it possible to follow this series while using just a python idle
love it thanks!
Nice Tutorial But where do I have to write the Code?
thanks for these great efforts can you create a playlist that contains all parts
Ya, should be on the channel now, updating as I add new bits.
Can we create a drowsiness detection andriod app with help of this series ?
i have a problem in the part of GPU Growth , i think because of the tensorFlow version and tried another modern versions but the same error that it doesnt recognize any GPUs on the actual machine
were you able to solve it?
I also faced the same problem
Please help me
@@sanjidahmazumder863 no
When I run the first Model(...) row I get the error "NameError: name 'inputimage' is not defined". please help
thank you.
Did you find any solution for this problem? If yes, please share.
can this face recognition app be used for commercial purpose in an organization
Once I started making directories
I am getting file exists error
Can anyone help me to solve this
I'm getting the same issues. So, what did you finally do to resolve the issue? Please share. It would be a great help for me. Thankyou!!
I experienced an error in the from matplotlib import pyplot as plt section
Please help!
what was the python version you had used for this project?
do you got the answer ?
can we recognize multiple faces at the same time ??
can u help me with the setup of the environment
You just jumped right into it. No guidance on where you're importing whatever before you started. This started off with a confusion for me.
i'm bigenner in IA development should i start with this tutorial
is it to use it for 3D face recognition please
I've watched your "Getting Started with Python Deep Learning for Beginners" video and then started doing this project, but when I want to execute and use the facial recognition app it gives me this error: "ERROR: Could not find a version that satisfies the requirement tensorflow==2.4.1 (from versions: 2.12.0rc0, 2.12.0rc1, 2.12.0, 2.12.1, 2.13.0rc0, 2.13.0rc1, 2.13.0rc2, 2.13.0, 2.13.1, 2.14.0rc0, 2.14.0rc1, 2.14.0, 2.14.1, 2.15.0rc0, 2.15.0rc1, 2.15.0, 2.15.1, 2.16.0rc0, 2.16.1)
ERROR: No matching distribution found for tensorflow==2.4.1"
I also faced the same... Can you please help me?
@@swarupsahu8560 I gave up on the project, then started working on google colab and chatgpt that way it was way easier to solve compability problems etc.
Can a beginner can follow this tut ?
is virtual environment required for this!!
I have the following error code on colabs and I wonder if you can tell me what is wrong, please:
!pip install tensorflow-gpu
Collecting tensorflow-gpu
Using cached tensorflow-gpu-2.12.0.tar.gz (2.6 kB)
error: subprocess-exited-with-error
× python setup.py egg_info did not run successfully.
│ exit code: 1
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
Preparing metadata (setup.py) ... error
error: metadata-generation-failed
× Encountered error while generating package metadata.
╰─> See above for output.
note: This is an issue with the package mentioned above, not pip.
hint: See above for details.
In the further parts of this series, we take up 3000 image pairs for training. But won't it overfit a siamese network based model? Also, we were supposed to make a one-shot model which would take only one image (or probably very few images) for training the model and then for verification too. But here we take as much as 3,000 image pairs for training and 50 images for final verification. Kindly clarify a bit over these matters.
Hey, are you following the tutorials end-to-end? If yes, could you please share what should I do when if I'm coding for creating the directories, it is showing that they already exist.
I tried experimenting with system environment variables, but unable to figure out the actual discrepancy.
Also, the images and everything that my webcam is capturing is getting stored but it's not clear in which path!!
Kindly help!
@@manasisingh294 Ig you can change all the paths and directories in the ipynb files (if you just want to run the project) and if you follow all the tutorials step by step, then you can create directories as you want. But if you are facing some problems, I would suggest follow what Nicholas is doing.
@@geek_ash Well, thanks for the reply!!
I did follow the video and executed all the required steps but still no progress.
can u make this project using ANN?? please
I had an erro while installing the first dependency (!pip install tensorflow==2.4.1 tensorflow-gpu==2.4.1 opencv-python matplotlib) i've already checked the requirements to installed and i have all so if you had the same problem i currently looking for a solution.
were you able to solve it?
Please help me in this@@sanjidahmazumder863
Same problem i face now
You don't mention your error, but you shouldn't have the exclamation point in front of the command
Do you have a plan dl+webapp??
Yup, will be a Kivy app!
now it is showing ERROR: Could not find a version that satisfies the requirement tenserflow-gpu==2.4.1 (from versions: none)
Try with Python 3.7.3
非常喜欢你的视频。
i am having python 3.9 version so I am not able to install tensorflow version 2.4.1 ....is there any solution for it ?
same problem.....did you found any solution?
Hi Mr Nicholas , is it possible to wrok with more 1000 faces effectively ?
Verify 1000 faces? I'm doing it for a single face but yep, the advantage of this model is that you can train to verify a bunch!
@@NicholasRenotte i love u man
@@NicholasRenotte It's great how you explain everything so nicely.. thanks a lot.. but can you please explain how to work with more than 1 face?
@@jaydattpathak4965 ya, will probably do another quicker series on doing it for multiple people!
@@jaydattpathak4965 ya, will probably double back with a vid on it. I didn't realise that would be the ask!
sir how can i instal kernel face id like u ???
Thank you.🙂
How about a model using YOLAC++ for real time object detection. Hope you see this comment!
Request to build deep fake Detection Model as well.
who else was bored during covid and decided to do this LOL
Hi, I have a question. I haven't idea of Face id folder. do I create it in local folder in our computer?
Hi bro can we connect?
@@nocopyrightmusic0031hi bro can we connect??
Hey, I love your videos and was trying to follow along, but for some reason, tensorflow won't detect my GPU despite reinstalling both the original version and the gpu version multiple times. do you have any fix for this?
Hello Nicholas Renotte,
From your this video (th-cam.com/video/06TE_U21FK4/w-d-xo.html&lc=UgybAvzcbfF4RvWJTj54AaABAg.9R3kZbbuxsV9R8MRuJWJzm&ab_channel=NicholasRenotte)
I have a question related to angle measurement, I want to get the angle for bowling action analysis... For that I need to calculate an angle between Shoulder, elbow and hand wrist... Form this video the angle is continuously measured from the camera but I want to calculate angle only when user perform a bowling action.
For the ICC rule, the angle is measured such that first we calculate angle once when user arm is parallel to ground (angle1) and when he throughs the ball (angle2) , after that I need to subtract the angle2 from angle1....
Please guide
Thanks
Could apply a pose detection model over the video first to detect when a player is in bowling then apply the angle estimate. Take a look at the body language decoder vid for how to do the first bit.
@@NicholasRenotte Thank you i have done it with your help and some research ❤️
Tensorflow is saying I have 0 gpus
But I do
Same, did you find a solution for this?
Thank feck i found you
this is unreal
Speak Slowly please.