?? Work at Pickup Points � Flexible schedule

Build an AI/ML Tennis Analysis system with YOLO, PyTorch, and Key Point Extraction

Code In a Jiffy 371 views

Download Video

Comments

I'm not able to fine-tune the tennis ball for yolo_inference py at around 19:30 Please help I was locally able to start the training process but to do in colab uploaded to drive and did the steps but was unable to train
Comment from : @im_simarjeet

can not download the dataset from roboflow : (
Comment from : @JIAYIYAN-k1j

Can anyone kindly teach me how to download the author's YOLOv5 Dataset as in 12:5013:26, because no matter how I clicked YOLOv5 button I can't see any download pop up message Even when I clicked Download Dataset button, nothing happens either Please help I tried it as I logged in two different accounts For the rest of other people's dataset I don't have any problem downloading
Comment from : @brownsquare77

how can we apply into camera application??
Comment from : @dustenobrial3943

Hi, one of my biggest problems after creating a Tennis video is to cut out the dead time during a match from the video How can I use this method to achieve that?
Comment from : @VishwaSeneviratne

very helpful video
Comment from : @hugohaus3961

How can we add in other shot type info like forehand/ backhand, volley , groundstroke ?
Comment from : @jan-lucako2042

I’m getting moduleError in roboflow part can you please help me sir
Comment from : @akashjayakumar6070

Can you do one for badminton?
Comment from : @TheBestGuy345

btw I instead of relying on player height and keypoints , you can try holographic projection , its not accurate but it works pretty good and ball position is quite pretty
Comment from : @iWillRun_K

1:32:06brbrif the keypoints are movingwhat we have to change?Can you explain it
Comment from : @hariharan29122

Hi, if there are any experienced ML specialists could i get some help with my questions?brbr1 Why has the author decided to use Yolo v5 instead of V8/V11br2 Why was the model tuned on 640x640 instead of 1280x1280 (as the tennis ball is small)br3 Why wasnt yolo also used for keypoint detection?
Comment from : @cristi4577

Very helpful ❤
Comment from : @PraveenKumar-dl3tu

is it possible to detech badminton shuttle with YOLO ?
Comment from : @iWillRun_K

when i train in the pycharm, it's report "ValueError: zero-size array to reduction operation maximum which has no identity"? why i download dataset from the roboflow,i think there is no problem in the dataset
Comment from : @hrz-kg5zf

can i do this entirey on google collab my ,PC is a potatoe, it can't even run IDE properly with 4gb ram
Comment from : @iWillRun_K

can you call yourself a pro after you complete this project?
Comment from : @cordz5274

After downloading the zip dataset from drive, it seems not possible to unzip it I try to download and unzip the file from local and colab too Anyone helps me with this please? Thanks
Comment from : @nhatanh4410

Why can’t I use another video as the input? It gets the error player_heights[player_id] Can anyone help me please?
Comment from : @mikhaaldyn112

I am very grateful for this video, which outlines the key procedures of YOLO deployment for an interesting project I've benefited a lot and also helped me finish a ping pong ball spin measurement project quite quickly: youtube/1c2kbc4hWuo
Comment from : @yuanhu6031

Hi, great works! I don't understand why you use pickle; better affidability?
Comment from : @nicolapomiato1737

Wow, amazing project sir! As a tennis fanatic and ML enthusiast, I consider myself very lucky to have found a project like this on YouTube Beautiful work, and I especially admire your teaching Comprehensive, slow and articulate Thanks a lot
Comment from : @florisbokx

Works on one video Try it on another and the ball goes the same way as in the first video
Comment from : @jon9066

i am not able to get the weights
Comment from : @hokagegaming5590

wow i really appreciate all the explanations How do you learn all this to just throw it out there with memory?! nicei would love to get with you on including more ai feedback with stats and even analyze how a play might have missed a shot, summarize ALL of that and give feedback to the playerchat?
Comment from : @RunnerProductions

works flawlessly thanks for taking the time to help sharpen our skills
Comment from : @emkbacon

fantastic , that's what we need brhey can you add more vedios on buildnig a web app machine thing?
Comment from : @Jafar801

Hi great video, Will this work for a paddle project?
Comment from : @jesuslopez6873

If I make this code for the tennis video you gavewill it work for other tennis videos too?
Comment from : @cutieeeeeeeee-d5j

Thanks dude But do you know what this project is for? Like, is it real world applicable? What use does this have IRL? Is this completely unique or an already existing thing? I'm so curious!
Comment from : @cutieeeeeeeee-d5j

Hey man that's awesome I also want to know that can it detect everything for other input video? If so can you provide links of videos with similar views?
Comment from : @Shan_____

19:36 how did you get that yolov5l6 file in colab do i need to download it
Comment from : @Abhishekkumar-hh5gk

Hi Thanks for your videos!!!br drivegooglecom/file/d/1UZwiG1jkWgce9lNhxJ2L0NVjX1vGM05U/view
Comment from : @titusfx

Hey,brbrAmazing video I am wondering if I can use your code and implement the project by myself and try to host it in some website Will definitely give you credit
Comment from : @ishanpurohit9762

Amazing video! Any ideas on how to detect bounces on court?
Comment from : @catalinawiedmaier2749

Abdullah, thank you for a wonderful tutorial, I have to get in touch with you cheers
Comment from : @ishiakugwamna1170

great video, can we do this in real time, and can we use this model in mobile app to test it in realtime? please answer it
Comment from : @AshutoshTitoria

When I am training the court keypoints on google colab, each step is taking 1 min and each epoch takes 80 mins, i am unable to train completely, in between the process, cell execution just terminates and i am unable to complete it, anyways, 20hour computation would not be possible in colab, Help me please
Comment from : @ABHISHEKKUMAR-l8i1u

I changed the input video and got an index out of memory error (probably due to the number of frames) How do test for other tennis videos ?
Comment from : @kanwarkelide8672

ha
Comment from : @haz-sh1tq

sir we want more such great big projects on object detection and nlp this was so unique everything was so well explained
Comment from : @shreyojitdas9333

how much time you spent on the making of this video?
Comment from : @batrasi

This is literally great quality of content out there, you made it easy to understand everything Thanks a lot
Comment from : @alpalimbachiya6377

Brother is underrated
Comment from : @IDKdk587

hi! can I request, can you make it into an application that can analysis the game while recording
Comment from : @Youfoundmej

Nice stuff, but noticed that the model is not robust enough to handle various footage of tennis when I try locally (especially for ball and court detection)
Comment from : @ismarapuandi1552

Can anyone share the ball prediction model!!!brbrI have tried to trained but it doesn't give required results
Comment from : @KhushPatel-x2n

Thanks for the video ! Is the technology precise enough to detect line calls ?
Comment from : @keneydy99

34:07 How do we decide the mean and std values while normalizing here?
Comment from : @TwilightCoder008

Swingvision needs to contact you guys, their app needs help in the android, Windows development area
Comment from : @graphguy

What are all prr requisites required to make such a project ?
Comment from : @abhinavsharma1175

I've been blessed by the youtube Algorithm I hit the jackpot
Comment from : @ramkumarsethupathi

It remains a mystery to me how you got such good results in training the court_keypoints model With the same code everyone gets a totally different loss descent! Do you have any idea of what worked better for you?
Comment from : @matteomainetti2581

My trackers are not right in the lines and the positions that they should be at On one side of the court they are fine but on the other it is skewed
Comment from : @Adh-Saksham

amazing can this be used for table tennis as well??
Comment from : @lalamax3d

can i do this similarly on yolo v10?
Comment from : @csrbharadwaj

SwingVision: 🤬
Comment from : @VisibleMRJ

great job sir! but were can i find the input videos am stuck in this
Comment from : @nadahammami750

please, give your vs code settings and installed plugins
Comment from : @ЮрийДоронин-н1ц

PAPER WHEN?! ❤
Comment from : @narangfamily7668

excellent job thank you very much can you please make more projects i like the way to present
Comment from : @farestabib6858

player tracker 51:07brball_tracker 1:20:48 brcourt keypoints 1:27:34brball tracker interpolation 1:48:00brchoose and filter players 1:59:00brBall Hit 3:03:00brPlayer pos to MiniCourt pos 3:23:00
Comment from : @umaiskhan6278

amazing project, just completed ittook me a week ;)brit was little hard but ultimately worth it!brGreat video :)
Comment from : @Sid-proxi

This is amazing Will this code work on other random youtube tennis videos?
Comment from : @yamumsyadad69

for train point key detection its take too much time cann anyone give me the model file
Comment from : @shubhasheeshkundu9933

I am literally copying you, but still at 1:46:13 few points (0, 4, 6, 1, 2, 5 ) are not at all on their respective position
Comment from : @Sid-proxi

Is anyone also struggling with downloading data for keypoint extraction ? Seems like the wget line is not working properly It downloads corrupted zip file with 3kB
Comment from : @jcobsky4852

can you provide the models pt files
Comment from : @elirana1000

can you do this for soccer?
Comment from : @paulpolizzi3421

nice tutorial
Comment from : @iasror

Abdullah, thank you so much for your hard work in this video I learned a lotbrbrYour teaching style is unrivalled I easily followed and understood everythingbrbrThank you, please keep up the good work
Comment from : @mh-ll

Love the idea I used another tennis video that I downloaded online, but it doesn't work with your code Should I be aware of any specifications regarding the video?
Comment from : @lanceschultz6154

Hello, when i try to use the code for a different video than the one you used as input it gives a key error in the convert_bounding_boxes_to_mini_court_coordinates function I don't know if you can help me with that 😕
Comment from : @goncalogomes1944

I believe this project is suitable for a newcomer
Comment from : @Trump-wr7ym

I loved this video! Thanks Would you make a video for analyzing a golf swing? I have to do a software engineering boot camp capstone, and I would love to incorporate this AI/ML model
Comment from : @lauratanguyen

Just stumbled upon this video by chance, and boy am I grateful!brAs a beginner to the world of ML, I think it would be amazing if you could make some beginner friendly project tutorials (I know a lot of other people would appreciate it too)brSomebody in the comments said something about an mnist digit recogniser? Would be cool to get a video on that!
Comment from : @wallisCodes-ig5lh

1:27:38 detect court keypointsbr1:48:00 interpolate_ball_positions functionbr1:56:00 Draw frame number
Comment from : @muonudom1566

bro this so diffeclet functionbrhow
Comment from : @cv_1960

Thank you!!! It's an excellent video I must say, no one teaches such detailed content on YouTube It's incredibly beneficial for students in engineering and technology I do have a few questions though Is this applicable in real time, such as by using a webcam?
Comment from : @JhonnyTech-ne9dt

bro please post a video how to set up vs code for ML and DSbr
Comment from : @FEELGOOD1820

Fantastic and helpful! 💯
Comment from : @pilau

Wow! Very cool and not super complex project Thank you for yolo introduction!!
Comment from : @Тима-щ2ю

There is only data_valjson but no data_trainjson file for some reason in the keypoints zip archive you downloaded with !wget
Comment from : @Тима-щ2ю

Is it possible to do this for table soccer?
Comment from : @Crz1985

This projects are insane, please continue the series, it has been very helpful
Comment from : @Amnskary

Can you make a video with trackiung basketball
Comment from : @RonanBorja

This is a great video Thank you so much
Comment from : @mwaqze

Analyze table tenis
Comment from : @mariozdk

I was about to go to bed Finding this video got me so excited that I'll probably be up another hour Can't wait to watch this tomorrow
Comment from : @boogersincoffee