Video is good and simple explanation made me to understand things. But i would like to know how you have made this Video and Presentation merging. look kind of cool to me. Please do let me know.
Good and helpfull video, but actually the Integral Image example is wrong. The formula is good, but the calculation is wrong. The purple areas sum is 9 2+5+2=9 You can combine it from the Integral Image where the A is acutally the top,left corner of the desired (purple) image section, what is 8. The D is always the origo of the desired image what is starts from the bottom, right corner IN the box, which is 46. The C is one left from the corner left border which is 13, and the D is one up from the upper right corner outside the box also which is 32. So based on the formula is: [A:8] [16] [24] [B:32] [C:13] [[23] [36] [D:46]] i(x',y') = S(A) + S(D) - S(B) - S(C) i(x',y') = 8+46-32-13 i(x',y')=9
One of the simple and best presentation on Facial Recognition that I have seen.
You are seriously a savior made it much easier to understand
Very nice and Helpful !
What about Part 2 ??
man special thanks to this explanation and demo easy and just perfect much respectt i wish u upload more thnxxxxxxxxxx a billion time
Is there a part II yet?
super and very use for my project. thank you!!!
how about if i want to sum the area of diagonal line haar features? does it have the same procedure?
hello sir i wana made eye detection system with respberry pi can it possible ? plz send idea and code...
Part 2, please !!!
Thanks you ,so much Tom!!
nice Work done brother ..
nice work
excellent ! clearly got the concept..Thanks a lot
hey . have a matlab code about this face detection method?
Great Job!
Video is good and simple explanation made me to understand things.
But i would like to know how you have made this Video and Presentation merging. look kind of cool to me. Please do let me know.
Good stuff. Thanks.
Summed area tables are just another one of those mathemagical things that I can't even imagine how they originally came up with.
This is perfect, thanks
Good and helpfull video, but actually the Integral Image example is wrong.
The formula is good, but the calculation is wrong.
The purple areas sum is 9
2+5+2=9
You can combine it from the Integral Image where the A is acutally the top,left corner of the desired (purple) image section, what is 8. The D is always the origo of the desired image what is starts from the bottom, right corner IN the box, which is 46.
The C is one left from the corner left border which is 13, and the D is one up from the upper right corner outside the box also which is 32.
So based on the formula is:
[A:8] [16] [24] [B:32]
[C:13] [[23] [36] [D:46]]
i(x',y') = S(A) + S(D) - S(B) - S(C)
i(x',y') = 8+46-32-13
i(x',y')=9
thanx for this post.. i was confused myself.. needed some confirmation
5:54 -- "Get the feature out of me!" :-)
Part 2, please !!!
hello sir, i wana made face recognition system with windows7 can it possible? please snd description nd code.
+madhuri khandelwal Try using Opencv-3.0.0. CascadeClassifier is the class you need. There are many blogs with code as well.
Go through my video, may be it is useful to you: ruclips.net/video/_BauBMxhJNs/видео.html