YOLO-NAS: Introducing One of The Most Efficient Object Detection Algorithms

Поделиться
HTML-код
  • Опубликовано: 10 янв 2025

Комментарии • 17

  • @jairajsahgal7101
    @jairajsahgal7101 Год назад +1

    Thank you

  • @thiagarajamuralidaran1371
    @thiagarajamuralidaran1371 Год назад +1

    Thank you

  • @Александр-б5б5ю
    @Александр-б5б5ю Год назад +1

    you make very cool videos!

  • @iyshwaryakannan6677
    @iyshwaryakannan6677 Год назад +1

    Hi sir, where can I find the YOLO NAS colab notebook to work on

    • @LearnOpenCV
      @LearnOpenCV  Год назад +1

      Hi, you can find the notebook here: github.com/spmallick/learnopencv/tree/master/YOLO-NAS_Introduction

  • @vishalmogal8839
    @vishalmogal8839 9 месяцев назад

    Hi Sir, Can you please make a video on YOLO world model? How it is different from other YOLO model with Use case for yoloworld model.

    • @LearnOpenCV
      @LearnOpenCV  9 месяцев назад

      Yolo World is a zero shot object detection model with its YoloV8 backbone to extract image features along with a shared vocabulary embeddings prompted by the user.
      This avoids the need to manually annotate images unlike traditional Yolo models.So when a user prompts with the desired class it's converted to vocab embeds,the box head shared with text contrastive head helps to find the object embeddings fusing text and image features
      Earlier Vocab embed models are transformer based backbone requiring heavy compute and slow wherein Yolo backbone models are known for lighweight and fast inference.
      For more on this checkout:
      docs.ultralytics.com/models/yolo-world/

    • @vishalmogal8839
      @vishalmogal8839 9 месяцев назад

      Thank you so much

  • @readbhagwatgeeta3810
    @readbhagwatgeeta3810 Год назад +1

    Can anyone please how to save the model after training on custom dataset, so that I don't have to train again and again for inference in different type of videos

    • @LearnOpenCV
      @LearnOpenCV  Год назад

      All the model checkpoints are saved in their respective "experiment-name" directories.

  • @sandecomx
    @sandecomx Год назад +1

    awesome

  • @AmitKumar-hm4gx
    @AmitKumar-hm4gx Год назад

    but how is a v5 model better than v7 ??

    • @LearnOpenCV
      @LearnOpenCV  Год назад

      YOLOv5 and YOLOv7 have different authors and developers.

    • @AmitKumar-hm4gx
      @AmitKumar-hm4gx Год назад

      @@LearnOpenCV I thought v7 is also from ultralytics…

    • @LearnOpenCV
      @LearnOpenCV  Год назад

      Check this repo for more info: github.com/WongKinYiu/yolov7