Ever wonder what your dog is thinking?
pursuit.unimelb.edu.auOne of the things I like about my cat (Eleanor Roosevelt, a one eyed Maine Coon that sat in the local SPCA for 8 months) is that I don't try to understand her. She does her thing, mostly involving naps and head scratches, and I just admire her calm and general contentment. Whatever your dog or cat is thinking, I would suggest you just leave it alone and appreciate the life those animals live that seem so contented. Who wouldn't want that?
I'll disagree, for my dog at least. It is fun to watch the gears turn in his head, sometimes I see the conclusion he'll get to before he gets there. And sometimes, since I know him so well, I know where his 'dog logic' will take him with his flawed world view. It makes a kind of internal sense, when you don't see the big picture. It's fun and made me realise how far AI is from even replicating a dog's language-less thought process.
It also makes me think of alien species who might view us the same way.
It has a practical side too, our guy usually slinks off to the couch for naptime after dinner but a few months back he walked to the office and just stood in the doorway staring at us. We could instantly tell by his expression that something was wrong. He turned out to have a grass seed lodged in his armpit (Which the vet did not find! I had to remove it myself the next day).
It is fun to just watch him sleep and dream, though.
> In the future, we have talked about analysing a pet’s body in addition to its face in order to improve the accuracy.
I think that's vital. Most of "Dog language" and even more of "Cat language" consists of body language. Things like pose and tension in relation to the rest of the things in the area at the moment. Dogs and cats are gifted with the ability to "speak human" enough to be understood tolerably well, but their language is demonstrably more complicated.
And many emotional displays are transient, like stress induced yawns in dogs (I note stress isn't in their top 5 list but it's common during training and learning new things).
I get that this is mostly for fun, but isn’t there a huge problem here, in that training the algorithm first requires human interpretation of the animal’s emotions? Which is a huge uncertainty in the first place?
Ah yes, I am sure the authors are aware of it. And i am sure they tried to ask animals first and then settled on human interpretation as second best option.
I woke up my dog to take a picture of her and see how this AI does. It ranked her as Unsettled (100%). Yep.
My poodle and I are both in the process of waking up. I got her attention to snap a pic, and it labeled her "Unsettled" too. Yeah, that's about what I feel before coffee, or in her case, fetch.
Tried this on a few photos of my dog - one of her half asleep, lying on her back and clearly very relaxed.
It said she was angry.
To be fair, it mis-identified her, and she is also a mixed-breed Asian dog so I'm not entirely surprised.
But it brings up questions for the future - what happens when AI is wrong but used as proof of something?
> what happens when AI is wrong but used as proof of something?
False positives everywhere. There was an article on HN a few weeks back about cops picking up the wrong guy, flagged by facerec. Even though the system's printout had a huge warning, "This document is not a positive identification", the detective still had the density to say, "So I guess the computer got it wrong, too."
https://www.npr.org/2020/06/24/882683463/the-computer-got-it...
> But it brings up questions for the future - what happens when AI is wrong but used as proof of something?
You bring up a fair point, but I don't think AI is special in this regard. Imperfect sources of information synthesis (humans, naive algorithms, etc) are used all the time. Hopefully, AI should simply add one more data point from which to understand a given situation.
People, presently seem much less willing to challenge what a computer says however.
I'm wondering if Little Britain's "computer says no"[1] sketch is somewhat foreshadowing here. It's certainly a situation I've ran in to in several banks (globally).
> But it brings up questions for the future - what happens when AI is wrong but used as proof of something?
I worry that the pace at which things like this will get packaged as “absolutely reliable, scientifically proven” technologies to all sorts of consumers will be far greater than the pace at which our language catches up with abstract ideas with which nuances can be widely communicated.
So an app on your phone, looking out through a camera on your glasses, could issue a security alert for very angry or scared people or animals in view. It seems a natural for Chinese style surveillance nets. Or it could be a learning tool or a prosthetic for the autistic. Real time AI assisted seduction/sales/politics/negotiation might become practical. How about an app for "reading the room".
People with the skills to read the subtleties of others' emotions have a great advantage. This tech could make some of that skill downloadable from the app store. Using it to tell if your cat is mad at you seems like the least of it.