Up front:Theres little evidence to believe thatso-called emotion recognition systemsactually work.
However, theres no ground-truth when it comes to human emotion.
Basically, nobody can tell if another person is gay, or aggressive just by looking at their face.
you’re able to guess.
And you might be right.
40% off TNW Conference!
What you might interpret as upset, might just be someones normal expression.
What you might see as gay, well.. Its impossible to train a computer to recognize emotions because computers dont think.
They rely on data sets labeled by humans.
Researchers dont train AI to recognize emotion or make inferences from faces.
They train AI to imitate the perceptions of the specific humans who labeled the data theyre using.
That being said:Creating an emotion recognition engine for animals isnt necessarily a bad thing.
From these, we infer the emotional states of animals in real time.
We covered something similar in the agricultural worlda while back.
Basically, farmers operating orchards can use image recognition AI to determine if any of their trees are sickly.
But AI can stare at trees all day and night.
AI for livestock monitoring is a different beast altogether.
According to the researcher, yes.
But according to theresearch: kinda.
The paper makes claims of incredibly high accuracy, but thats when compared against human spotters.
Maybe the next chef doesnt like it, maybe nobody but that chef likes it.
Unfortunately, the main reason why this matters is because livestock that lives in relative comfort tends toproducemore.
Or women, typically, rate pain higher when observing its perceived existence in people and animals.
Which bias do we train the AI with?
you could read the whole paperhere.
