Scale Invariant Feature Transform 1 (Feature Detectors)
ฝัง
- เผยแพร่เมื่อ 20 พ.ค. 2024
- In this video, I have discussed the SIFT detector in detail.
Slides:
drive.google.com/file/d/1VP9r...
Sources.
Visual interaction and explanation of the algorithm.
weitz.de/sift/index.html
Blog on SIFT
towardsdatascience.com/sift-s...
Kristen Grauman’s slides
www.cs.utexas.edu/~grauman/cou...
Professor Ajit Rajwade's slides
mafiadoc.com/scale-invariant-...
Distinctive Image Features from Scale-Invariant Keypoints (SIFT Paper)
by David G. Lowe
www.cs.ubc.ca/~lowe/papers/ij...
Sunflower images
Photo by fotografierende from Pexels
Photo by Ana Arantes from Pexels - วิทยาศาสตร์และเทคโนโลยี
![ใจเสาะบนเกาะนาคี - อ๊ะอาย สกุณาพร เสียงวิหค [Official MV] | ค่ายเพลงเป็นหนึ่ง](http://i.ytimg.com/vi/jbcmGQkfyOg/mqdefault.jpg)
Very well described and shown, the best!
you are gifted man! this was so much fun :)
Hey buddy. You have a really nice explanation style and it comes from the fact that you have a very in depth understanding of the subject matter. This was my first time learning SIFT and I understood it completely. Good job. If you continue to post technical content like this, ill definitely subscribe. Hope your interest in computer vision stays the same!
Thank you so much for the feedback. Yes I will keep uploading videos .. 😊✌
Great explanation Sir.
excellent explanation!! Thankyou
Thank You so much very well explained
Is there any implementation available for this from scratch i.e. without using the library?
Awesome explanation
very nice explanation thanks sir..
Thank you man for this video. It's really helpful
@Kaleb Omar nice try guys you are so believable omg..
Great work man
Thanks for the nice video. I have a doubt in the step 'scale-space extrema detection'. For an octave: (considering 5 different scales of images created using Gaussian blur), we would be having 4 resulting DoG images from the previous step. So it's understandable to compare pixels from the 2nd DoG image with its neighbors from the 1st and the 3rd DoG images. Similarly, we could compare pixels from the 3rd DoG image with the 2nd and the 4th images. But how about the pixels in 1st and the 4th (topmost in that octave)? With whom should those be compared? Or we just consider only those in the middle (2nd and 3rd from the four DoG images from the previous step!!)
We only consider those that are in middle . For more better understanding you can see the description i have put a link of visual interaction and explanation of algorithm please check that.
Loved the way you explained it. Thanks a lot.
I have one question. In scale-space extrema detection, do we need to always compare the middle pixel of the second(intermediate image)? I don't understand that part.
Thank you for your feedback. 😇
To answer your question, it's not necessarily the second image it can be any image from the second image to second last image. Basically the pixels should have 26 neighbours in total. 9 above 9 below and 8 in the same plane.
Can you please make a video on OLPP?
Wow quite lucidly explained! At 28:05 how does the transpose of the inverse multiplied with the original matrix become identity matrix? Inverse multiplied with the matrix itself gives an identity right?
Anyway thanks for the video!
Thanks for the feedback. Since Hessian matrices are symettric tye transpose of inverse is equal to the inverse matrix.
I don't understand why we do scale space extrema detection and not just space extrema detection. In a previous slide, you show a 1D example with convolution with the laplacian of Gaussian, but there are not several values of sigma in that slide...
If you see the slide titled (Coming to the point) you can see that the concept works only when the size of the blob is similar to that of the sigma value of the laplacian. Hence its important to serach within a range of different sigma values
Very nice explanation
How to contact you. I need bit of clarification on SIFT
Can I use haris corner detector and then SIFT descriptor far face?
impressive man ,
thumbs up
WOW
why this is a derivative of guassin?
I did not understand your question. Is your question why is the laplacian of gaussian a derivative of gaussian ?
Gaussian simply blurs the image (or we can say cancels the white noise). Besides, derivate of gaussian determines the changes in pixel values/edges. Therefore, for detecting edges it's necessary to use the derivative of gaussian.
讲的挺好的，就是印度口音听着有点别扭