Face recognition in images and video with Python (opens in new tab)

(pyimagesearch.com)

281 pointszionsrogue8y ago28 comments

28 comments

27 comments · 9 top-level

tjoff8y ago· 8 in thread

A bit unrelated but I'm in a need for some sort of cat deterrent that will spray water at all cats except mine. Also it would be nice if humans were exempted as well.

I haven't done the legwork myself yet, but would this be an appropriate project to check out for this? (As for identifying my cat I've been thinking about alternative solutions such as rfid necklace or something but it is quickly getting past 'quick hack').

scardine8y ago

Check out the talk "Militarizing Your Backyard with Python: Computer Vision and the Squirrel Hordes"[1] from PyCon 2012. He wired a water canon to a camera able to tell squirrels from birds:

> Has your garden been ravaged by the marauding squirrel hordes? Has your bird feeder been pillaged? Tired of shaking your fist at the neighbor children? Learn how to use Python to tap into computer vision libraries and build an automated sentry water cannon capable of soaking intruders.

Previous discussion here: https://news.ycombinator.com/item?id=3710107

[1] http://pyvideo.org/pycon-us-2012/militarizing-your-backyard-...

rhplus8y ago

A Bluetooth LE device similar to Tile[1] might be a cheap and reliable signal source in a package small enough to attach to a collar. If you had two Bluetooth receivers on either side of your camera's visual field, you could compare signal strengths during motion and positively identify your cat (assuming feline motion detection is a solved problem).

[1] https://www.thetileapp.com/

zmmmmm8y ago

It might be easier to cheat and put a bluetooth or similar beacon onto your cat to suppress the water spray than try and distinguish your cat from others?

tjoff8y ago

One of the goals is to create a safe place for my cat, on the porch or something. And still be able to fend off neighbor cats. She is getting old and has had fights on the porch which makes her scared to go out.

So I guess it depends on how easy it would be to limit the scope/direct the receiver to only activate when she is on the stairs which is probably the optimal target range for the porch (and not activate when she is lying just beside the receiver).

Thanks so much everyone for the answers! Can't believe I forgot about the squirrel presentation. My cat is all black and I'm not sure how I'd go about distinguishing it. A really colorful necklace perhaps, but it would have to be quite ungraceful to not be covered in fur when watched from the front/back (a second camera 90 degrees from the first could perhaps help).

In the QA in the squirrel presentation someone also mentioned OpenTLD, which seems to be superseded by CMT ( https://www.gnebehay.com/cmt/ ). Worth a look.

I hope I can find the time to experiment on this.

Tiki8y ago

I don't think you'll find a 'quick hack' but what you're asking is definitely possible as a project.

Check this video out for ideas(guy used ML/servo controlled squirt gun to get rid of squirrels from his bird house) https://www.youtube.com/watch?v=QPgqfnKG_T4

ourcat8y ago

This older article from the same author might help you : 'Detecting cats in images with OpenCV' - https://www.pyimagesearch.com/2016/06/20/detecting-cats-in-i...

manifestsilence8y ago

This would be amazing. Please do a DIY blog if you figure this out. :)

subcosmos8y ago

I find it amazing that we live in a world where such profound tech is purposed for such tasks as this :)

kruhft8y ago· 3 in thread

Can it track eye movements? A cheap eye tracker would be awesome so all the window's i'm looking at are in focus.

zionsrogueOP8y ago

This method is just for face recognition. If you're interested in tracking eye movements take a look at face detection + facial landmarks:

https://www.pyimagesearch.com/2017/04/03/facial-landmarks-dl...

And then techniques to localize the center of the eye:

http://thume.ca/projects/2012/11/04/simple-accurate-eye-cent...

taeric8y ago

Sounds like a great idea. I would be a little worried about how often I actually am not looking at the window I am typing in. Though, I suspect there are obvious patterns.

kruhft8y ago

I think it would be easy to train for novices.

1 more reply

symisc_devel8y ago· 2 in thread

We released a C/C++ computer vision library (Python binding to follow) earlier this month which let you perform face recognition at Real-time from IoT or Raspberry PI devices. The library is dependency free, cross platform and should compile fine on most modern architectures with a C compiler.

https://github.com/symisc/sod

Djvacto8y ago

Speaking as someone with limited and theoretical knowledge of computer vision, does CV usually overlap with Optical Character Recognition (OCR)?

I read through the readme and don't recall any references to identifying text, so I was wondering if and where a separation between OCR and CV might exist in the development process?

symisc_devel8y ago

Yes, CV is always the first pass before character extraction. It envolve a lot of image processing routines including morphological operation like dilation, noise removal, etc. and finally blob detection[1]. All of the preprocessing routines are implemented in SOD.

[1]: https://sod.pixlab.io/c_api/sod_image_find_blobs.html

JulianWasTaken8y ago· 2 in thread

Nice.

I did something simple here awhile back with a CFFI wrapper for OpenCV.

http://tech.magnetic.com/2015/11/real-time-facial-recognitio...

vstuart8y ago

Ditto :-) http://persagen.com/about/victoria/projects/cv_facial_identi...

drhoffma8y ago

But Jurassic Park theme!

nickserv8y ago· 2 in thread

Great article. I'm working on a project to identify buildings, and it seems this technique would apply. Does anyone know of a good buildings dataset, or a pretrained network? 3 million images is no easy task to find... Thanks!

kevinali18y ago

Aerial images?

nickserv8y ago

Sorry, forgot to mention this would be for street level recognition. I don't think aerial views would work for that application, perspective issues notably.

ebrewste8y ago· 1 in thread

Pyimagesearch is great. I'm just a tinkerer / hobbyist in the computer vision space, but this is a site I come to time and again to get up to speed on topics that just seem harder elsewhere.

andyidsinga8y ago

true that - a lot of very practical examples and posts on this site ..and python makes them a joy to play with!

edit: fwiw - I've been playing with motion detection / background subtraction https://www.pyimagesearch.com/2015/05/25/basic-motion-detect...

leblancfg8y ago

I automated half of my first desk job by writing this Python CLI tool, which wraps the OpenCV `detectFace` method:

https://github.com/leblancfg/autocrop

Pyimagesearch is a great repository of everything OpenCV related, great to see them on the front page.

wcrichton8y ago

If your use case is archival (running this on many videos stored on disk) as opposed to realtime (running on a live video feed), we built a research tool called Scanner [1] for maximum performance in offline video analytics, e.g. see our face detection example [2].

[1] https://github.com/scanner-research/scanner

[2] https://scanner-research.github.io/scannertools/#face-detect...

turowicz8y ago

The article is very similar to readme.md and samples from the github repo of face_recognition.

j / k navigate · click thread line to collapse