Dear B-A,
Cornell Lab of Ornithology is developing a project called Merlin. This aim of this project is to develop software that enables computers to identify birds from images. See http://dev.nabirds.org/Web/Tools/ImageUpload/pages/about.php
Given the advances in human biometrics, I imagine that computer ID of birds is quite doable. Wonder how long it will take for the technology to appear in handheld devices? After all, how many of us say, 10-15 years ago, would have thought that an all singing, all dancing field guide in a smartphone type device would be possible?
An interesting space to watch.
Cheers,
Carl Clifford ===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line) to: birding-aus-request@vicnet.net.au
http://birding-aus.org ===============================
Hi Ian,
Thanks for the post, very interesting. It is going to take me some time to read Ryan Farrell’s papers and to digest them. The field of computer vision is exciting and very rapidly evolving with so much potential to bring about interesting changes to our lives. I have been a nerd with a wide range of interests all my life and still have trouble keeping up with tech, mainly due to the gobsmack factor. Things have changed so much since my first boyhood foray into tech (a crystal set radio made from scratch using a Galena crystal and a tuning coil made from wire scavenged from a model A Ford ignition coil, way back in the depths of the last century). I still find myself staring at my iPod Touch and shaking my head over what I can do with it.
To answer your questions: (i) no; (ii) none of any great size, though there are plenty of excellent collections of images out there in the hands of individuals and (iii) ? maybe.
This is a space I am going to keep under observation.
Cheers,
Carl Clifford
PS, welcome to Aus. We can always do with a bit more depth in the gene pool.
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
Dear All,
I’m new to birding-aus and have some comments questions about this.
In my non-birding life I work on computer vision, having just moved to Adelaide from Oxford two months ago. I know Pietro Perona who is the computer vision lead and have had some discussions with him about the project. I have also birded once or twice at conferences with one of the postdocs associated with another group who collaborate, Ryan Farrell. If you are interested in the state of the art then as well as checking out Pietro’s site, also look at Ryan’s paper from last year’s International Conference on Computer Vision at this web page http://www1.icsi.berkeley.edu/~farrell/
The state-of-art is some way from being able to identify birds in general images, or even from nicely posed photographs. Most human biometric stuff that actually works relies on quite controlled conditions for acquisition of the imagery (e.g. passport photo recognition), including the pose and lighting. These are two key factors that make identifying a bird in a photo or video really tough. Furthermore it is an open research question to what extent one needs to program expert knowledge into a machine recognition system for this kind of subordinate categorization task, versus being able to learn it from training data. The latter has met with some stunning successes in recent times, and in terms of general object class recognition computers can do reasonably well on 10s or sometimes 100s of distinct classes, but there is no clear answer as to how far this can be pushed when trying to distinguish very closely related subordinate classes.
This area of so-called “Fine-grained Visual Categorisation” is one that interests me and I would love to think there is a will within Australia to do something similar to the Cornell/Caltech, so it would be great if there is any advice from this list that people could offer. In partic, (i) is there any equivalent to Cornell (or the Edward Grey Institute in Oxford) in Australia that would have an interst in engaging in such a project; (ii) what databases of photographs exist and how can they be used/accessed/etc; (iii) would there be any will from the birding community at large to participate in markup of a database with a view to building an identification tool for Australian and/or Australasian birds?
Cheers, Ian
John,
I am not sure that the University of Wisconsin is quite up to that. A big job. They may be happy to lease the algorithms though. They are a pretty public spirited lot, having given the world the technology of using UV light to produce vitamin D and also the anticoagulant Warfarin.
Perhaps some entrepreneurial type in Australia might wish to take up the torch for an Australian version.
Cheers,
Carl Clifford
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
Thanks Carl:
You’re right – someone will come up with regional editions down the track if the licensing fees are not too steep. The hard part will be getting all the audio samples – seems you need dozens of reliable source recordings of each call of each species to make a comparison file for the software to reference for greater accuracy…..could be challenging in regions lagging behind North America and Europe in the field recording area.
Cheers,
John
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
Looks interesting – I hope they expand it to be worldwide coverage!
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
Hi Dave,
A program that identifies bird calls has already been developed in the US, but only for research as yet. I last heard about it at the Australasian Ornithological Conference in 2011, but they were up to about 90% accuracy on ID from memory.
Cheers, Chris
Chris,
There is an app for North American species (WeBIRD) being launched next year. See http://www.news.wisc.edu/19882
Cheers,
Carl Clifford
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
Better than me.
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
I guess he became more interested in the sound of one wing flapping.
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
This one, which appears to be Australian, claims it can identify calls with 95% accuracy: http://www.soundid.net/
I assume that’s if you’ve got the appropriate recordings to compare to.
Peter Shute
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
Then we need that linked to a direction finder that moves your bins to point to the right place in the tree and birding skills will be almost superfluous!
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================
Interesting – the very first birding database I used was called Merlin. I used it for a couple of years then the guy who developed it told his users he was going off to be a Buddhist monk and would no longer support the product so I moved on….
And then of course we need one that identifies calls – there are services that will identify music I know (you play a sample you heard on the radio and it tells you what the music was) but this would be a bit harder I suppose – or maybe it already exists?????
===============================
To unsubscribe from this mailing list, send the message: unsubscribe (in the body of the message, with no Subject line)
http://birding-aus.org ===============================