Raspberry Pi LESSON 60: Finding and Tracking Faces In OpenCV with Haar Cascades

แชร์
ฝัง
  • เผยแพร่เมื่อ 12 ธ.ค. 2024

ความคิดเห็น • 28

  • @AyraWinla
    @AyraWinla ปีที่แล้ว +1

    I don't like coffee, I still like the occasional cat video and I unfortunately don't own a Raspberry Pi, but I still find this video series incredibly helpful. It's teaching me the basics of OpenCV and image recognition in a very easy to understand way, and I've been following along with my Windows PC perfectly well with only minor code changes required and everything works perfectly. Besides the pan-tilt part obviously, but the logic is clear and simple enough.
    I've been doing amateur robotics stuff for a while with things like the Pico and doing things like a 'radar' with a laser displaying the returns on a screen. I've been sitting on some ideas for trying to grab stuff for a while and was thinking of maybe dipping my toe with color recognition to see the possibilities and wasn't really sure on how to get started... This video series was absolutely perfect and gave me the answers I needed. Definitively enough information to get started on grabbing a clearly defined object with known dimensions. Those videos really kickstarted me in a good direction.
    A huge thank you! I'll still watch the occasional cat video, though.

  • @pralaymajumdar1206
    @pralaymajumdar1206 11 หลายเดือนก่อน

    Sir, you taughtus this technique in Jetson nano lesson 53 & 54 in different procedure... It's just outstanding.
    You make me an Engineer in different level. You r awesome sir.

  • @eugeneamoakwah278
    @eugeneamoakwah278 ปีที่แล้ว +2

    Really love the way you teach and encourage us to do things on our own. Looking forward to use the pi to develop any project I want. Thanks.😇

  • @imadjawad4408
    @imadjawad4408 ปีที่แล้ว +2

    Hi Paul, very exciting lesson! thank you, looking forward to the next lesson.

  • @cbrombaugh
    @cbrombaugh ปีที่แล้ว

    I AM LEGEND! I am able to see the face and 2 eye rectangles, but it is hard to see all 3 together for more than a couple seconds. My excuse for the poor performance is that I have poor lighting in my lab. I am only getting 1 frame per second. The slow speed is probably because am running my 8 GB Pi 4 with VNC from my old laptop. When I installed the pan-tilt unit, my Pi became permanently unable to produce HDMI output. I guess I will have to break down and buy a new Pi at the atrocious $170 price that the gougers are charging these days.
    Thanks for all these great lessons Paul. This AI stuff is really neat!

    • @charlotteswift
      @charlotteswift ปีที่แล้ว

      @Charles Brombaugh I replied to a very old comment of yours that said your HDMI output had stopped working. I don't know whether you got it. Could you possibly have plugged the HDMI cable into the wrong HDMI output on the Pi? Sometimes it can be something really simple. I somehow managed to screw up the output on my Pi 4 but I can't remember how I got it working again.

  • @suryatejasv5870
    @suryatejasv5870 ปีที่แล้ว +1

    Thank you Paul ! Keep the amazing content coming out way

  • @keithlohmeyer
    @keithlohmeyer ปีที่แล้ว

    Nice video! I will be watching from the sidelines with my weak Pi 3. Thanks Paul for all you do

  • @426F6F
    @426F6F ปีที่แล้ว

    I'm on the Arduino projects right now, but I can't wait to get here! Thanks Paul for all you do!

  • @scottwait3585
    @scottwait3585 ปีที่แล้ว +1

    Thanks Paul!

  • @qzorn4440
    @qzorn4440 ปีที่แล้ว

    Hi Paul. I have some interesting results off of the Orange Pi 5 16GB Rockchip RK3588S 8 Core 64 Bit Single Board Computer, 2.4GHz Frequency with 5V4A TypeC Supply Amazon $150
    Debian-Bullseye-OS Python3.9-OpenCV using 2 different USB webcams, which turned out to be about the same, running Lesson-60 Haar_Cascades and Htop results:
    800 x 600; 24 FPS; 73 degrees C; 1.8 GHz; 8 cores 72% 1280 x720; 23 FPS; 76 degrees C; 2.0 GHz; 8 cores 73% Anyway I thought I would share the progress.
    The image quality was very nice and it detected face and hand movement quick. The OPi-5 works well with an USB WiFi/BT dongle. 😎 Thank you.
    Also FYI there is a new WiFi/BT version: Amazon $174 Orange Pi 5B 16GB Rockchip RK3588S 8 Core 64 Bit WiFi6,BT5.0 Single Board Computer

  • @arguedas02
    @arguedas02 ปีที่แล้ว

    Thank you Paul, what a Great lesson…

  • @Matlab_Minds-vikas
    @Matlab_Minds-vikas ปีที่แล้ว

    you are always best sir

  • @ade1963
    @ade1963 ปีที่แล้ว

    brilliant - thank you

  • @HA-wo3we
    @HA-wo3we ปีที่แล้ว

    You are the man!. Can I use openCV with pi zero w? do you thing this little thing has enough power?

    • @paulmcwhorter
      @paulmcwhorter  ปีที่แล้ว

      It should work. Just might be a little slow. I think I played around with it before.

  • @claviclvl3
    @claviclvl3 8 หลายเดือนก่อน

    I'm not able to find the haar files on my RPi 5 and running on bookworm. Was there a significant change in how cv2 works?

  • @FourthWayRanch
    @FourthWayRanch ปีที่แล้ว +1

    why is there no information about how to control one microcontriller with a signal from another microcontroller? I have to connect a pin to ground from another one with a 3.3V output signal. Can't we just do this with a transistor or something other than resorting to using an actual relay?

    • @keithlohmeyer
      @keithlohmeyer ปีที่แล้ว

      No relay needed. If you are just controlling in one direction make the receiving board pin an input with a pull up or down and the transmitting board as output. I would put a 220 ohm resistor in the circuit just in case you get the coding wrong. For more robust 2 way comms use serial rx/tx 2 pins for short distances or add a couple rs485 breakouts and you can go 100's of feet. I have had that setup running between 2 Arduinos in my one acre greenhouse transmitting temperature humidity data. It has been running rock solid 24/7 for 2+ years.

  • @-darnasek-1872
    @-darnasek-1872 ปีที่แล้ว

    Hi! i'm getting a (-215:Assertion Failed) !empty() in function 'detectMultiScale' I'm running on a picamera v3 noir. What could be the issue?

    • @-darnasek-1872
      @-darnasek-1872 ปีที่แล้ว

      It seems it was a path problem. I solved this by removing the . in front of the whole path.

  • @jameslewellen150
    @jameslewellen150 ปีที่แล้ว

    Here is the homework for this lesson. th-cam.com/video/KSOA_Qla1F8/w-d-xo.html

  • @DrDave327
    @DrDave327 ปีที่แล้ว +1

    Another great lesson, Paul!! Here is my solution to the homework assignment:
    th-cam.com/video/MJ2-SfzcPTE/w-d-xo.html

  • @hammadahmed3409
    @hammadahmed3409 8 หลายเดือนก่อน

    Homework th-cam.com/video/JWh3Rpv3cP8/w-d-xo.html