If you are getting an error while installing requirement “lap==0.4.0” then try installing visual studio C++ complier as shown at 1:10:10. Let me know if that solves your issue.
i don't have this problem just with lap but also the specific version of opencv-python and ultralytics. I still can't solve it.pls i need some help cause this is my second course iwan not to skip too because of the same problem that i found with pycocotools
59:23 You can also get the classnames from the results variable: class_names = results[0].names print(class_names) The only difference is that it will be a dict with index keys as ID numbers and that's probably what you want but can be turned into a list with: class_names = results[0].names print(list(class_names.values()))
59:20 YOLO already have a name attribute: results = model.predict(stream=True, imgsz=512) # source already setup names = model.names for r in results: for c in r.boxes.cls: print(names[int(c)])
fantastic work Murtaza, thank you! This made me think of autonomous drones: if you combine Yolo object detection with the Tello camera input and make it go forward as long as it detects objects in the room and stop it, when no longer detecting, you would have a sort of lidar for free...
For anyone struggling with the requirements.txt, make sure you have Python 3.10 or above. I was trying to run it on Python 3.9, and it was saying there was some hardware error. I completely deleted and reinstalled python 3.10, and everything works perfectly. Besides that, this was a highly informative course on the YOLO algorithm and its applications. Thank you Murtaza.
This was the fix for me as well. I tried originally installed 3.12, and had tons of issues with the requirement packages. I even tried 3.11, and that was better, but still issues. Only after uninstalling everything and going with JUST 3.10, it worked!
How beautiful if I had seen your video earlier. I not only learned technology, but also practiced English. Thanks a lot for your efforts and I will continue to follow you.🥳
Dude, i want an object detection / reader / scanner that works over a visor or glasses in real time. Could be very very basic at first but over time it would be an incredibly useful tool for the everyday human.
Excellent video Murtaza. I think you should have figured out by now that in the x,y,w,h system, the x and y refer to the centre of the bounding box and not the top left. That's why you had the rectange starting from the centre of your torso in the initial webcam example with cvzone package.
This course help me for my learning project. ❤ Thank you sir. I need more specific counting for every class object. Could you make new video tutorials about it sir. Thank you again
Ow yeah i was waiting for this thank u very much and for people who coulnd't install lap library like me in chapter 7 murtaza install visual studio packets it will fix the lap library problem
Its was very helpful. Thank You Man. Im staisfed there is broccoli in classes because they are very helthy. btw. i found reasonable to set terminal shell to cmd.exe by default. Setting>Tools>Terminal > Shell path: switch to cmd.exe
in 26:42 you said -in your website there is a zip file . but i didn't find any zip file.please help me.also searched in object detection course.but nothing found to download.
Hi . Can you please help me with the installation of the nvidia drivers because I am not able to install the nvidia drivers but have successfully installed both toolkit and cudnn. It's showing that the NVIDIA installer cannot continue because no nvidia GPU is detected on your system. This graphics driver could not find compatible graphics hardware.
Good work. Murtaza!. I would like to point one thing in car counter project. Instead of list for tracking counts, you might need to consider dictionary or a set for unique ids.
I would like to express my sincere thanks for your time, effort, and willingness to share your expertise with the online community. Your contributions have made a significant impact on my learning journey, and I am truly grateful for that. Once again, thank you for your exceptional work and for going above and beyond to educate and inspire others. I look forward to exploring more of your content in the future.，you are the best one
Hi . Can you please help me with the installation of the nvidia drivers because I am not able to install the nvidia drivers but have successfully installed both toolkit and cudnn. It's showing that the NVIDIA installer cannot continue because no nvidia GPU is detected on your system. This graphics driver could not find compatible graphics hardware.
@@Piyushraj-ry1kz what python version you have? i had issues while installing lap package and I downgrade the python version and it all works fine after that
hello Murtaza. firstly, great video. you have explained everything se well. in the poker hand project, you have labelled the Ace as 14, which will work fine for the upper straights. however for the straight Ace to 5, we need it as 1. so we need to write an extra line of code stating that if my list contains, 14, 5, 4, 3, 2 this is also a Straight and not a High Card. otherwise all's good. once again thank you for this amazing video!
I don't know how to thank you but, you really helped me as I wanted a detailed tutorial on how to implement this step by step. Thank you, and keep up the good work.
@@user-rn1hu3ij4d I think yes , try the code but I think the difference that as an example - the first project - the video and detection will be slow .
Хочу сказать тебе что ты молодец. Я однажды делал видеопокер и отдаю себе отчёт что значит кодировать логику работы автомата. Мне понравился твой подход к решению задачи. Молодец!
Nice tutorial. The car count is not working 100% as expected - several vehicles change id number close to the red line. So there is a risk that a vehicle can be counted twice. You can see an example of this in the video around 2:17:29 where the red bus and the white truck beside it change ids just before the red line. Not sure if this can be corrected with the Sort parameters?
please any help regarding this error mgRegion = cv2.bitwise_and(img, mask) cv2.error: OpenCV(4.5.4) D:\a\opencv-python\opencv-python\opencv\modules\core\src\arithm.cpp:214: error: -209:Sizes of input arguments do not match) The operation is neither 'array op array' (where arrays have the same size and type), nor 'array op scalar', nor 'scalar op array' in function 'cv::binary_op'
Awesome video. I've been watching your videos nonstop for the past 2 days and I've learned tons. Thank you so much! A small tip for the last part of chapter 6 (Yolo with webcam), instead of hardcoding the class names you can simply access them using the model object like so: model.names[cls] Hope this helps :)
Hi, amazing tutorial! I have two question about the poker project: 1. Let’s say I have two different detectors, one for numbers (rank) and the other one for suit. How can use it detector to avoid to make a new “poker card detector”? 2. It’s possible to make an AI model to classify the hand given the detected cards? I mean, to avoid the traditional code in the last part of the tutorial. How? Thanks again!
