I found Computer Vision: Algorithms and Applications really good. You can download it for free (for personal use) at http://szeliski.org/Book/.
Ask HN: What are the best resources to learn computer vision?
11–20 of 59 posts
Re: Ask HN: What are the best resources to learn computer vision?
#12I started by getting a webcam or two and trying out various projects: marker tracking (made an optical IR pass filter and tracked an IR LED with two cameras), object segmentation (e.g measure geometry of certain-colored objects). Measure the speed or count the number of cars passing by your street. Try to implement an OCR for utility meter. There are lot's of applications you can train yourself in, and I guarantee th…
This won't get you an interview, let alone a job.
Re: Ask HN: What are the best resources to learn computer vision?
#13There is a processing library with opencv bindings here:
https://github.com/atduskgreg/opencv-processing
Re: Ask HN: What are the best resources to learn computer vision?
#14I once created a list of 20 problems in CV. If you solve them in order, you will know basic CV. Here goes: https://gist.github.com/abhinai/b6eebecb4d19c57cfb1ee64c2b53...
A good online reference is: http://opencv-python-tutroals.readthedocs.io/en/latest/py_tu...
Re: Ask HN: What are the best resources to learn computer vision?
#15Does anyone know if tech like OpenCV is used at companies developing their own "computer vision" product, maybe at Tesla? Or do they build their own technology from scratch which isn't available to public domain? Or do they say fork OpenCV and build upon it and heavily modify as OpenCV could be seen as 'outdated' technology.
Disclaimer: Never worked with any technology related to Computer Vision, just a bloodboy beginner Python programmer.
Re: Ask HN: What are the best resources to learn computer vision?
#16For a full course, Nothing beats CS231n: Convolutional Neural Networks for Visual Recognition http://cs231n.stanford.edu/ by Andrej Karpathy et al.
Also, for a general and high level introduction to neural networks, I wrote a Learning Deep Learning in Keras http://p.migdal.pl/2017/04/30/teaching-deep-learning.html, focusing on visual tasks.
Re: Ask HN: What are the best resources to learn computer vision?
#17cs231n by Andrej Karpathy : http://cs231n.github.io
This is a great resource. I give it to people who need to learn about convolutional neural networks.
However let's keep in mind that the field of computer vision is much vaster than that. Deep learning approaches have been very successful at solving problems in computer vision, but not all of them and not without drawbacks. I believe any course on classic computer vision will give him more insight as to what challenges computer vision aims to solve, how, and what approach might solve what problem.
Re: Ask HN: What are the best resources to learn computer vision?
#18For a full course, Nothing beats CS231n: Convolutional Neural Networks for Visual Recognition http://cs231n.stanford.edu/ by Andrej Karpathy et al. Also, for a general and high level introduction to neural networks, I wrote a Learning Deep Learning in Keras http://p.migdal.pl/2017/04/30/teaching-deep-learning.html , focusing on visual tasks.
I did this course, but couldn't finish all the assignments. Loved it. Please note that this is a convolutional neural networks course, not computer vision as such. From what I know computer vision encompasses a variety of non machine learning based algorithms, which are not covered in this course.
Re: Ask HN: What are the best resources to learn computer vision?
#19OpenCV and http://www.pyimagesearch.com/ disclaimer: not related to any of these
2nd for pyimagesearch. The author (not me) is a prolific and dedicated blogger who really wants to share his knowledge. Dude has a 'bootcamp' as well
Re: Ask HN: What are the best resources to learn computer vision?
#20Does anyone know if tech like OpenCV is used at companies developing their own "computer vision" product, maybe at Tesla? Or do they build their own technology from scratch which isn't available to public domain? Or do they say fork OpenCV and build upon it and heavily modify as OpenCV could be seen as 'outdated' technology. Disclaimer: Never worked with any technology related to Computer Vision, just a bloodboy begi…
Used to work on Tesla Vision / Autopilot Vision. They used Caffe, were switching to Tensorflow when I left, but might be moving to Caffe2 now.
Usually no OpenCV on successful products. Facebook Ads has dedicated research engineers implementing their real time photo analysis algorithms.