Thanks a lot! This was very easy to follow along indeed. In case anyone's teachable machine page isn't training, try using it on a browser with no added extensions, and clear cache.
python[27190:320561] WARNING: AVCaptureDeviceTypeExternal is deprecated for Continuity Cameras. Please use AVCaptureDeviceTypeContinuityCamera and add NSCameraUseContinuityCameraDeviceType to your Info.plist. WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1712824494.653097 320561 gl_context.cc:357] GL version: 2.1 (2.1 Metal - 86), renderer: Apple M1 INFO: Created TensorFlow Lite XNNPACK delegate for CPU. Traceback (most recent call last): File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/ops/operation.py", line 208, in from_config return cls(**config) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/depthwise_conv2d.py", line 118, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/base_depthwise_conv.py", line 106, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/layer.py", line 263, in _init_ raise ValueError( ValueError: Unrecognized keyword arguments passed to DepthwiseConv2D: {'groups': 1} During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/Users/oorjamishra/Desktop/hand ASL tracker/test.py", line 12, in classifier = Classifier("model/keras_model.h5", "model/labels.txt") File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/cvzone/ClassificationModule.py", line 29, in _init_ self.model = tensorflow.keras.models.load_model(self.model_path) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/saving/saving_api.py", line 183, in load_model return legacy_h5_format.load_model_from_hdf5(filepath) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/legacy_h5_format.py", line 133, in load_model_from_hdf5 model = saving_utils.model_from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/saving_utils.py", line 85, in model_from_config return serialization.deserialize_keras_object( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/serialization.py", line 495, in deserialize_keras_object deserialized_obj = cls.from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/models/sequential.py", line 326, in from_config layer = saving_utils.model_from_config(
Hello! Sorry for bothering, but perhaps can you help me with this tutorial? I can't make it work and it's been like a whole month since I started lolol. Thanks in advance!!!
This is without a doubt the best sign language recognition tutorial out there. Believe me when I say this because I have seen so many. Thank you for saving my career. God bless you.
Absolutely marvelous! Everything worked as demonstrated, even the errors, and your resolutions on your errors worked for me too. Thank you very much sir. Could you show us how to do the learning on PCs instead of Teachable Machine? Also, what is the difference in the different formats you can save your model in, where and how are they used and how to convert from one format (say h5) to another, say tfjs. Thanks again for taking your time and putting in a lot of effort to make this kind of very quality video! Cheers from Nairobi, Kenya!
Your next tutorial should be how to program AI to understand ASL words, phrases, and sentences. It'd be multithreaded where one thread understands letters, while another is always keeping a log of the last few gestures in real life to understand how to connect them together as a phrase.. Something like that. I'm not an expert with AI. I believe that is kind of how live closed captioning works on TH-cam.
why my test code is not working its giving output like train code and still saving while i press "s" and not making prediction. still training output like imgwhite is appearing with "Right " lable not "A" or "B"
hello sir i have seen the entire tutorial that was great and the way u explain every step is awsome but i am facing a problem in sabing image what should i do
Never saw Chinese drama this was 1st one soo well made and especially this part is soo emotionally attached....chai xaoqi and Fang leng wonderful love from India❤..
Great video! Very clear on how to solve this problem. Is it possible to give an example on how to train for video classification to identify movements, like PySlowFast? Tks
This video is my big wish,thank you for that. But you used prepared functions from cvzone. Can you do this manuel model that you composed for maybe eye directions that a human stared.
Do we have to install some other libraries before tensorflow . I use python 3.7 and pycharm doesnt recognize tensorflow i guess i installed tensorflow but it doesnt detect the letters after that . It only identify the hand but not the trained characters Does anyone know how to fix this?
Hi Sir, first of all, outstanding work. I want to know one thing while storing the dataset of hand gestures, we are also getting the landmarks points but if I don't want these hand landmarks then what do I need to change in the code. Thanks
TNice tutorials was super helpful, dude! I got the tutorial version of soft soft just to get a taste, and after figuring it out I decided to purchase the
For people who want to fix error at th-cam.com/video/wa2ARoUUdU8/w-d-xo.html ``` if (x > 0 + offset and y > 0 + offset and w > 0 + offset and h > 0 + offset): cv2.imshow("Croped", imgCrop) ``` Juste add this when you showing your cropped image
Hi thanks for your input Did you install any other libraries or dependencies before installing tensorflow (like CUDA ORcudnn ) my system doesnt recognize tensorflow Appreciate if you can help me
@@shubharthaksangharsha6248bro please help getting error python[27190:320561] WARNING: AVCaptureDeviceTypeExternal is deprecated for Continuity Cameras. Please use AVCaptureDeviceTypeContinuityCamera and add NSCameraUseContinuityCameraDeviceType to your Info.plist. WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1712824494.653097 320561 gl_context.cc:357] GL version: 2.1 (2.1 Metal - 86), renderer: Apple M1 INFO: Created TensorFlow Lite XNNPACK delegate for CPU. Traceback (most recent call last): File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/ops/operation.py", line 208, in from_config return cls(**config) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/depthwise_conv2d.py", line 118, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/base_depthwise_conv.py", line 106, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/layer.py", line 263, in _init_ raise ValueError( ValueError: Unrecognized keyword arguments passed to DepthwiseConv2D: {'groups': 1} During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/Users/oorjamishra/Desktop/hand ASL tracker/test.py", line 12, in classifier = Classifier("model/keras_model.h5", "model/labels.txt") File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/cvzone/ClassificationModule.py", line 29, in _init_ self.model = tensorflow.keras.models.load_model(self.model_path) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/saving/saving_api.py", line 183, in load_model return legacy_h5_format.load_model_from_hdf5(filepath) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/legacy_h5_format.py", line 133, in load_model_from_hdf5 model = saving_utils.model_from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/saving_utils.py", line 85, in model_from_config return serialization.deserialize_keras_object( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/serialization.py", line 495, in deserialize_keras_object deserialized_obj = cls.from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/models/sequential.py", line 326, in from_config layer = saving_utils.model_from_config(
The system predicts the value correctly but multiple times for the same sign due to the speed in recognition. But how to take only one value for a gesture? or how to slowdown the recognition process or how to recognize only ones at equal intervals of time store the recognized values some where?
@Sahil Sawant yeah that part is easy, that's just how you detect 2 hands. But how do you get one bounding box for both hands? And then cropping it and everything
how to stop the program crashing when we zoom in the hands? Can't we add an if statement with regards to limiting the resolution displayed? Or what is the try except you're talking about??
sir why you don't use jupyter notebook or google colab to make projects most of the students do not have gpu in their computers for projects so they can't use pycharm so please can you use google notebook or jupyter notebook. beacause sometime project fails when we do the same project you taught in pycharm whe n we try to implement it in googlecolab.
I’ve cracked soft soft before twice as a kid. Never learned anytNice tutorialng to make anytNice tutorialng useful. Sad tNice tutorialng is a lot of my peers back then
Hi for those of you who are struggling and having issues when you move your hand outside the borders of the camera, there's a bug where you will get cv2 error. Use this code to fix it if imgCrop is not None and imgCrop.shape[0] > 0 and imgCrop.shape[1] > 0: **paste it below if aspect ratio if aspectRatio > 1: .........
I have a question I have a project and I really need it If you can help me or make video How create custom object detection using yolov3 or yolov5 and train it And export Weight file to use it with python and opencv with video or webcam. And Thank you❤️
This is so useful, actually I have trained model on all letters of alphabet as well as number 0-9 taking all 36 classes into account. But I am facing a little issue with the code. It would be better if you can help me out with the same.
@@abhaybhatt8007 these both signs are kind of motion sign so I initially created the sign by which it starts so J would be tilted I towards the little finger side and for Z I used tilted 1 on thumb side. But most importantly I had to change the entire algorithm and with the help of that I was able to achieve very better accuracy...
@@fizaeerum4216 exactly, what I felt was that it's the complexity issue and so I changed the algorithm to train the model... With the help of that I was able to achieve very better accuracy
Hi sir ! I saw the entire tutorial ,you not only taught us the concepts of sign language detection but also u taught the perfection. Thank you sir :)
it took me 6-7 days to run this program fully ........and thankyu so much for this video.....this videos is easy to understand for beginners
can you please send me the github link
did u get the link bro@@mounikatelagamsetti4788
Bro...can you send github link
Hii Mam/Sir I too started this project but it is not working so kindly guide me to complete this project Thank you
I started an identical project a while ago, but couldn't finish it ... After seeing this video, I finally finished my project.
Thank You
He brother i need your help regarding the project as i have same related project
How can I contact you? I also have the same project.
@@HELLO-cz4vm Through Linkedin
Bhai source code de do plz yr
hello, i need your help with this project, I am having an error, how can I contact you?
Thanks a lot! This was very easy to follow along indeed.
In case anyone's teachable machine page isn't training, try using it on a browser with no added extensions, and clear cache.
python[27190:320561] WARNING: AVCaptureDeviceTypeExternal is deprecated for Continuity Cameras. Please use AVCaptureDeviceTypeContinuityCamera and add NSCameraUseContinuityCameraDeviceType to your Info.plist. WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1712824494.653097 320561 gl_context.cc:357] GL version: 2.1 (2.1 Metal - 86), renderer: Apple M1 INFO: Created TensorFlow Lite XNNPACK delegate for CPU. Traceback (most recent call last): File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/ops/operation.py", line 208, in from_config return cls(**config) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/depthwise_conv2d.py", line 118, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/base_depthwise_conv.py", line 106, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/layer.py", line 263, in _init_ raise ValueError( ValueError: Unrecognized keyword arguments passed to DepthwiseConv2D: {'groups': 1} During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/Users/oorjamishra/Desktop/hand ASL tracker/test.py", line 12, in classifier = Classifier("model/keras_model.h5", "model/labels.txt") File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/cvzone/ClassificationModule.py", line 29, in _init_ self.model = tensorflow.keras.models.load_model(self.model_path) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/saving/saving_api.py", line 183, in load_model return legacy_h5_format.load_model_from_hdf5(filepath) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/legacy_h5_format.py", line 133, in load_model_from_hdf5 model = saving_utils.model_from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/saving_utils.py", line 85, in model_from_config return serialization.deserialize_keras_object( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/serialization.py", line 495, in deserialize_keras_object deserialized_obj = cls.from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/models/sequential.py", line 326, in from_config layer = saving_utils.model_from_config(
Please help get this type of error
Thank you sir, i have done my graduate project with the help of this video and your my god once again thank you
Hello! Sorry for bothering, but perhaps can you help me with this tutorial? I can't make it work and it's been like a whole month since I started lolol.
Thanks in advance!!!
hello, can you please help me resolve two errors in this project please ...........
The world not be the same without you.
This is without a doubt the best sign language recognition tutorial out there. Believe me when I say this because I have seen so many. Thank you for saving my career. God bless you.
hello, sorry for asking, but did you manage to get it to work? :C
Well, I did not try with C but I used other signs and they worked well @@kiaraagustinaalegreencinas9458
Absolutely marvelous! Everything worked as demonstrated, even the errors, and your resolutions on your errors worked for me too. Thank you very much sir. Could you show us how to do the learning on PCs instead of Teachable Machine? Also, what is the difference in the different formats you can save your model in, where and how are they used and how to convert from one format (say h5) to another, say tfjs. Thanks again for taking your time and putting in a lot of effort to make this kind of very quality video! Cheers from Nairobi, Kenya!
Really respect and appreciate people like yourself who give their ti and expert knowledge so others can benefit and enjoy soft
Your next tutorial should be how to program AI to understand ASL words, phrases, and sentences. It'd be multithreaded where one thread understands letters, while another is always keeping a log of the last few gestures in real life to understand how to connect them together as a phrase.. Something like that. I'm not an expert with AI. I believe that is kind of how live closed captioning works on TH-cam.
why my test code is not working its giving output like train code and still saving while i press "s" and not making prediction. still training output like imgwhite is appearing with "Right " lable not "A" or "B"
hello sir i have seen the entire tutorial that was great and the way u explain every step is awsome but i am facing a problem in sabing image what should i do
Never saw Chinese drama this was 1st one soo well made and especially this part is soo emotionally attached....chai xaoqi and Fang leng wonderful love from India❤..
bro what
@@albsjalbs2251exactly 😂
Great video! Very clear on how to solve this problem. Is it possible to give an example on how to train for video classification to identify movements, like PySlowFast? Tks
You taught us with very simple and clean. Thank You Sir..
This is the video I've been looking for! Thank you for your work!
This video is my big wish,thank you for that. But you used prepared functions from cvzone. Can you do this manuel model that you composed for maybe eye directions that a human stared.
Amazing tutorial! Works like a charm
know where to start a new soft and I didn't know how to switch from soft to . You are the best THANK YOU FOR NOT
Love your handtracking tutorials. Always tend to fascinate me and inspire my work in my job. I really love your videos Mr. Murtaza!
THANKS AAAA LOTTTT SIRRRR
Hey were you able to get this project to work?
Amazing sir ....with your video i get to learn so many things ...thanks alot once again for this video🎉🙏🙏
Very helpful video ,what if we want 2 hand images. what will be the code for it sir
Please make a video on sign language detection that converts ASL to sentences.
hence on here a month later. thanks for making feel normal again.
Very helpful, and surprisingly therapeutic
Thanks a lot sir. It's really easy to code and learn at the same time.
Which algorithm have u used ?
Ty I got my first divine because of you
Will stay tuned to your channal! Cheers!
great video i think i will switch to computer vision its so fun!! thanks .
Great tutorial, Just wondering does that detector module detect at larger distance ?
Thank you for using automatic code formatter. Btw, what is cvzone? Is it your own library?
It's his website name
Yes, I think it's his own library. We have to install that package.
It works! Thank you for your help!
totally agree with u mate! im kind of a newbie to soft producing here, well tNice tutorials video helps a lot and gives mo motivation to create
Very helpful and easy to understand, thanks a lot sir!
Hi can help me with the data using your hands. Need it for accuracy
Do we have to install some other libraries before tensorflow . I use python 3.7 and pycharm doesnt recognize tensorflow i guess i installed tensorflow but it doesnt detect the letters after that .
It only identify the hand but not the trained characters
Does anyone know how to fix this?
I have the same issue... so frustrating.
i am unable to install tensorflow. i think my pythn version is old. pls share a way to install tensorflow!
You are the best! Thank you!
Hi Sir, first of all, outstanding work. I want to know one thing while storing the dataset of hand gestures, we are also getting the landmarks points but if I don't want these hand landmarks then what do I need to change in the code. Thanks
TNice tutorials was super helpful, dude! I got the tutorial version of soft soft just to get a taste, and after figuring it out I decided to purchase the
Thanks for the beautiful explanation
May I know which algorithm have u used in this project?
Hi, firstly ı would like to say thank you for this videos. If we use two hand for our mother tongue's sign language, how can ı revize this project?
Which part of the code should i modify for data collection and testing for create a project using 2 hands
Thanks you man for sharing this stuff.
For people who want to fix error at th-cam.com/video/wa2ARoUUdU8/w-d-xo.html
```
if (x > 0 + offset and y > 0 + offset and w > 0 + offset and h > 0 + offset):
cv2.imshow("Croped", imgCrop)
```
Juste add this when you showing your cropped image
Thank You very much
Hi thanks for your input
Did you install any other libraries or dependencies before installing tensorflow (like CUDA ORcudnn ) my system doesnt recognize tensorflow
Appreciate if you can help me
Which part of the code should I put this? need help
@@dibyajyotigoswami2151 How did you fix it sir??
Thanks man, really easy and interesting project.
THANK YOU FOR THIS!!!
This is very helpful, thank you so much.
I just wanted to ask, will this work without nvidia cuda and cudnn?
Perfect 🤩 thanks
What system is this running on? is there a link to where pycharm and supporting software are installed?
tHANKS bUDDY fOR GREAT VIDEO !!
You're the best please make some arduino+ opencv too
he has already made it but that is not free
I know, I wish he could make some more free Arduino videos that will be greatt. I really love his videos @@himanshukulkarni9535
@@shubharthaksangharsha6248bro please help getting error
python[27190:320561] WARNING: AVCaptureDeviceTypeExternal is deprecated for Continuity Cameras. Please use AVCaptureDeviceTypeContinuityCamera and add NSCameraUseContinuityCameraDeviceType to your Info.plist. WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1712824494.653097 320561 gl_context.cc:357] GL version: 2.1 (2.1 Metal - 86), renderer: Apple M1 INFO: Created TensorFlow Lite XNNPACK delegate for CPU. Traceback (most recent call last): File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/ops/operation.py", line 208, in from_config return cls(**config) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/depthwise_conv2d.py", line 118, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/convolutional/base_depthwise_conv.py", line 106, in _init_ super().__init__( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/layers/layer.py", line 263, in _init_ raise ValueError( ValueError: Unrecognized keyword arguments passed to DepthwiseConv2D: {'groups': 1} During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/Users/oorjamishra/Desktop/hand ASL tracker/test.py", line 12, in classifier = Classifier("model/keras_model.h5", "model/labels.txt") File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/cvzone/ClassificationModule.py", line 29, in _init_ self.model = tensorflow.keras.models.load_model(self.model_path) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/saving/saving_api.py", line 183, in load_model return legacy_h5_format.load_model_from_hdf5(filepath) File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/legacy_h5_format.py", line 133, in load_model_from_hdf5 model = saving_utils.model_from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/saving_utils.py", line 85, in model_from_config return serialization.deserialize_keras_object( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/legacy/saving/serialization.py", line 495, in deserialize_keras_object deserialized_obj = cls.from_config( File "/Users/oorjamishra/miniforge3/envs/moon/lib/python3.10/site-packages/keras/src/models/sequential.py", line 326, in from_config layer = saving_utils.model_from_config(
This is very helpful for me.
Please what is the laptop requirements for this.
Thank you so much
thank u ur an amazing programer
The system predicts the value correctly but multiple times for the same sign due to the speed in recognition. But how to take only one value for a gesture? or how to slowdown the recognition process or how to recognize only ones at equal intervals of time store the recognized values some where?
thank you that's so helpful❤
Thank You For TeacNice tutorialng Us Brother
Thanks. Tried kasp last time.. so iobit is a better app it seems..
Can you make a video one sticker overlapping and skew detection using openCV and webcam?
I once did this with Google's nocode ML kit environment but it's better to have your own model.
great.. good job keep it up
Thank you, sir!
Do we need to resize the bounding box? cause our image on train data have same size.
Thank you so much sir 💖
thanks sir,you are great
Amazing video keep going ❤
What should I do if I want to use 2 hands instead of 1 hand?
@Sahil Sawant yeah that part is easy, that's just how you detect 2 hands. But how do you get one bounding box for both hands? And then cropping it and everything
hi sir ,the algorithm of this is style augmentation? right
THANK YOU SO MUCH
Great video 👌👌
hello i have a qst plz which method of machine learning you used?
bro i need help like what are the features are extracted by the model during training?
explain that
Amazing ✨✨
Sir, what method does this project use?
how to stop the program crashing when we zoom in the hands? Can't we add an if statement with regards to limiting the resolution displayed? Or what is the try except you're talking about??
During the Croping of the image I keep on getting this error "(-215:Assertion failed) size.width>0 && size.height>0 in function 'cv::imshow'"
Same, could you solve it ?
@@leticiabatista122 noo, I'm still searching to get it done...but not very lucky
change the camera to (0)
Number 1 is CNG
what should I add to detect two hands together?
Sir, how to display the accuracy
sir why you don't use jupyter notebook or google colab to make projects most of the students do not have gpu in their computers for projects so they can't use pycharm so please can you use google notebook or jupyter notebook. beacause sometime project fails when we do the same project you taught in pycharm whe n we try to implement it in googlecolab.
You don't actually need a GPU for this project.
All the computation can be done by the CPU, however depending on your CPU it may be slower.
@@B1NT0N I am talking about all the other projects not only this.
How about linking some of your video into Home Assistant for home automation!
This is amazing❤️
Thanks a lot
how i can remove points on hand that is detected in order to pass it to model without lines on it ??
sir I want to use an external camera connected via ethernet cable. How can I do this?
Sir you have used teachable machine to train the model. If i want to train the model in the code itself how can i do that? can someone help?
I’ve cracked soft soft before twice as a kid. Never learned anytNice tutorialng to make anytNice tutorialng useful. Sad tNice tutorialng is a lot of my peers back then
I am having issue with the keras model? Can anyone tell me in which version of keras and tensorflow the model is trained?
Awesome content
Thanks a lot :)
Hi can help me with the data using your hands. Need it for accuracy
hello sir can i please know how to do it for further alphabets please
How to solve this error: (-215:Assertion failed) !ssize.empty() in function 'cv::resize'
Could you solve it?
@@pedrovasquezmontilla679 bro i solved it
@@pedrovasquezmontilla679 if imgCrop.size >0:
ImgResize = cv2.resize(imgCrop, (wCal, imgSize))
else:
Continue
And also if that doesn't work ..try to use your smartphone as webcam because that will input more clear frames.
@@MuhammadUsman-hb6zkyes it's worked .
Thanks brother
Can you please make one for recognizing old video games and giving a price? Some old video games are worth alot
Hi for those of you who are struggling and having issues when you move your hand outside the borders of the camera, there's a bug where you will get cv2 error. Use this code to fix it
if imgCrop is not None and imgCrop.shape[0] > 0 and imgCrop.shape[1] > 0:
**paste it below if aspect ratio
if aspectRatio > 1: .........
can you explain it little bite more what to do?
Works well!! DANKEEE
Is it possible to detect 1 hand or 2 hands and body interaction?
I have a question
I have a project and I really need it
If you can help me or make video
How create custom object detection using yolov3 or yolov5 and train it
And export Weight file to use it with python and opencv with video or webcam. And Thank you❤️
Thank you Boss
Which editor is this?
This is so useful, actually I have trained model on all letters of alphabet as well as number 0-9 taking all 36 classes into account. But I am facing a little issue with the code. It would be better if you can help me out with the same.
Yeah me too I am also facing some problems, it doesn't recognise letters after q
How you Train J And Z
@@abhaybhatt8007 these both signs are kind of motion sign so I initially created the sign by which it starts so J would be tilted I towards the little finger side and for Z I used tilted 1 on thumb side. But most importantly I had to change the entire algorithm and with the help of that I was able to achieve very better accuracy...
@@fizaeerum4216 exactly, what I felt was that it's the complexity issue and so I changed the algorithm to train the model... With the help of that I was able to achieve very better accuracy
hello I am having eroor of hands, img = detector.findHands(img) after saving the model part . can you please help
^^^^^^^^^^^^^^^^^^^^^^^