Object tracking using YOLOv9 and ByteTrack | Ultralytics

Поділитися
Вставка
  • Опубліковано 1 жов 2024

КОМЕНТАРІ • 45

  • @afollowerofchrist5789
    @afollowerofchrist5789 6 місяців тому +1

    Thank you very much for the wonderful content as always. Just a very minor typo I believe from a previous YOLOv8 template. I believe you meant "# Load a pretrained YOLOv9c model" instead of "# Load a pretrained YOLOv8n model".
    Another question is why the model is so big compared to YOLOv8n? It's much more comparable to YOLOv8m, and though it may seemingly more accurate, the inference time sometimes doesn't make sense, as it is higher than that of YOLOv8 sometimes. Do you happen to have an explanation for it? Thank you again for the wonderful content!

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      Correct for typo - yolov9c instead of yolov8n . Thanks.
      Inference time - It might be due to the architecture of the model. I haven't explored much from that angle but will surely try to figure out the reasons and will cover in upcoming videos. Thank you 🙂

  • @ArunKumar-sg6jf
    @ArunKumar-sg6jf 27 днів тому

    custom object detecor how to do byte track

  • @norapap3163
    @norapap3163 6 місяців тому

    I love your videos, always super useful! I am trying to work with yolov7 and ByteTrack, but it always asks for yolox. Do you have any idea how I could use only Bytetrack for tracking? I have my detections values in a csv file I would read in.
    Thanks for your answear!

  • @soravsingla8782
    @soravsingla8782 6 місяців тому

    Hi Aarohi, your content is excellent and your channel is one of the best Artificial Intelligence channel but still not getting that much of response which your channel deserves. Hope you succeed

  • @RoyMathew-h1g
    @RoyMathew-h1g 6 місяців тому

    Hello ma'am. How can count the object in the frame while tracking. I think we have to count the track IDs right. Can you please provide me the code maam

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      Code related to this video : github.com/AarohiSingla/Object-tracking-Using-YOLOv9 and yes you need to count the track ids

  • @archishasingh7299
    @archishasingh7299 6 місяців тому

    Maam how do I save my training result when training using ultralytics to my google drive?
    I want to resume training from last checkpoint.
    Also i the dataset to be localised in colab's working environment

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      To resume the training just add --resume in your training command. It will automatically pick the last trained weights. Yes, you need to put the dataset in google drive.

  • @SagarLekhak
    @SagarLekhak 3 місяці тому

    Wish I was rich enough to donate you enough.

    • @CodeWithAarohi
      @CodeWithAarohi  3 місяці тому

      Thank you so much for your kind words! Your support means the world to me, whether it's through donations, likes, comments, or just watching my videos. I'm grateful to have viewers like you!

  • @NamNguyen.ee2
    @NamNguyen.ee2 6 місяців тому

    Thank you for your video. Please help me to know that how can I see file run after running these code on Google Colab?

  • @jayeshbaviskar9251
    @jayeshbaviskar9251 6 місяців тому

    Perfect... Was waiting for this❤ can you make a video on norfair tracking? Its latest version is amazing. Yolo v8 and norfair tracking

  • @AfnanBasheer-v2r
    @AfnanBasheer-v2r 6 місяців тому

    while training my model on custom dataset using yolov9 I'm getting cuda memory error , did u also get the error?

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      No, I didn't. You can reduce the batch size and then try

  • @sravanthmahendrada4611
    @sravanthmahendrada4611 6 місяців тому

    can you please make video on object detection and video summarization as they are not available in youtube

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      Great suggestion! Will surely do it.

  • @maryaliza709
    @maryaliza709 6 місяців тому

    Hi Aarohi, your videos are wonderful and very helpful for me. I'm wondering if you can provide a tutorial on how to train custom dataset such as wheelchair and kids and adults as pat of classes to detect and track? do you suggest using Google Open Images Dataset?

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      Gather images and videos that contain instances of wheelchairs, kids, and adults. Annotate all the images.
      Select a suitable deep learning architecture for object detection and tracking.
      Train model and use the trained weight file for detection.
      Regarding using Google Open Images Dataset, while it provides a vast collection of images with annotations for various object classes, it might not cover all the specific classes you mentioned.

    • @maryaliza709
      @maryaliza709 6 місяців тому

      Thank you so much for your help :)@@CodeWithAarohi

  • @noorahmadharal
    @noorahmadharal 6 місяців тому

    Ma'am also cover the inpainting algorithm and their application and image upscale with different deep learning techniques like SRCNN , SRGAN , ESRGAN and how can we train them on our custom dataset.

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      SRGAN tutorial: ua-cam.com/video/rVtJxfYVLrQ/v-deo.html
      SRGAN: ua-cam.com/video/KsNLxBvJBKo/v-deo.html
      I will make videos on the rest of the requested topics soon 🙂

    • @noorahmadharal
      @noorahmadharal 6 місяців тому

      @@CodeWithAarohi Thank you miss

  • @arpittalmale6468
    @arpittalmale6468 6 місяців тому

    Ma'am you have made any paid courses for deep learning and object detection ? Because I want to learn end to end project of object detection from you

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому +1

      Hi, I don't have any paid courses. I only teach through UA-cam videos.

    • @arpittalmale6468
      @arpittalmale6468 6 місяців тому

      @@CodeWithAarohi ok ma'ak thank you so much for your response

  • @58kartikaursang49
    @58kartikaursang49 6 місяців тому

    I want to detect the vehicale and also count. On the signal 🚥 using cctv camera , how to do it

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому

      I will cover in next video.

    • @58kartikaursang49
      @58kartikaursang49 6 місяців тому

      @@CodeWithAarohi i want to do a project using that so i need it please 🙏🏻 share the idea

    • @CodeWithAarohi
      @CodeWithAarohi  6 місяців тому +1

      @@58kartikaursang49 Check this video, you just need to provide the cctv as source : ua-cam.com/video/sqwSJsFdZeA/v-deo.html

  • @AmiSoS
    @AmiSoS 5 місяців тому

    a simple and understandable video. congratulations

  • @aneerimmco
    @aneerimmco 3 місяці тому

    Informative :)

  • @arnavthakur5409
    @arnavthakur5409 6 місяців тому

    Your videos are amazing ma'am

  • @Sunil-ez1hx
    @Sunil-ez1hx 6 місяців тому

    Simply amazing video👍👏👏

  • @No-Userrfound
    @No-Userrfound 6 місяців тому

    Thank you Mam

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 6 місяців тому

    Good