Calculate the Area of an object | with Mask R-CNN and Python

Поделиться
HTML-код
  • Опубликовано: 10 дек 2024

Комментарии • 15

  • @pysource-com
    @pysource-com  2 месяца назад

    🔥Learn how to build your own AI vision solutions: pysource.com/community

  • @somusundram1823
    @somusundram1823 2 года назад +6

    I achieved H x W x L (3D) calculation with IntelRealSense D4xx camera,combining opencv + math + mrcnn. Like you said before, a well trained model can give an accurate inference in a matter of +- 2cm tolerence!

  • @gplgomes
    @gplgomes 2 года назад +4

    Other easier way is to train a classification of Legos type, since there are only a few. So the classification find the type and associate to its dimension.

  • @muddassarjilani4637
    @muddassarjilani4637 2 года назад +2

    Great lesson, can we do it using detectron2, if possible kindly make a video, very thankful to you

  • @elifozcan9345
    @elifozcan9345 2 года назад +3

    How can I calculate the region for irregular images? Like brain tumor detection?

  • @VinhNguyen-bc2yu
    @VinhNguyen-bc2yu 2 года назад +1

    Can you make a video about realtime detection with colab?? I really need it

  • @waleedak6164
    @waleedak6164 2 года назад +1

    How can i export the masks with their areas into a csv file?

  • @marco29616
    @marco29616 2 года назад +1

    ¿if i want to see the class names what are the changes of the code i can do?

  • @aymanlamzouri123
    @aymanlamzouri123 2 года назад +1

    Can you do for us some excel python traitement courses

  • @aaabeenav7
    @aaabeenav7 Год назад +3

    The code is very expensive. Please make it affordable. THanks

  • @lucastavares7767
    @lucastavares7767 2 года назад +1

    Very good😃🇧🇷👏

  • @izurzuhri
    @izurzuhri 2 года назад +1

    please help i can't install on my notebook using google collab

  • @qaisadawi6144
    @qaisadawi6144 Год назад +1

    The code is not working now could you please help me?

  • @guilhermematos2340
    @guilhermematos2340 Год назад

    Why not use simple segmentation? That is, using only image processing, instead of having to train a CNN model?