China 'Smart Restaurant' Uses Facial Recognition To Make Meal Suggestions (techcrunch.com)
An anonymous reader quotes TechCrunch's report about Baidu's newest project:
The search giant sometimes referred to as the 'Google of China' partnered with KFC to open a new "smart restaurant" in Beijing, which employs facial recognition to make recommendations about what customers might order, based on factors like their age, gender and facial expression... image recognition hardware installed at the KFC will scan customer faces, seeking to infer moods, and guess other information including gender an age in order to inform their recommendation... And the setup also has built-in recognition, so if you're a return customer, it can 'remember' what you ordered before and suggest your past favorites.
Baidu has also worked on another KFC restaurant in Shanghai where the orders were taken by a voice-activated robot.
Baidu has also worked on another KFC restaurant in Shanghai where the orders were taken by a voice-activated robot.
Based on the roundness of your face.
Based on the roundness of your face and the profit margin. They don't care about your health.
Live today, because you never know what tomorrow brings
It's not really about helping you out with recommendations. It's about providing a novelty, seemingly personalized service as a way to sell more food. It also helps introduce new products and encourage customers to try different things, maximizing profit for the restaurant.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
No soup for you!
Great minds think alike; fools seldom differ.
If the algorithm is actually really smart, it will advise you to avoid all fast foods, including KFC.
Slashdot, fix the reply notifications... You won't get away with it...
Wonder how this could go wrong.....
Silence is a state of mime.
if China KFC is anything like the American version most people when they get to the checkout knows what they want.
Yes, and that is the problem they attempt to solve. The goal would be changing what people want towards what gives the most profit in the least amount of time, or new items that they want customers to sample at least once.
A little nudge, a little reminder of other items, but within what has a higher likelihood of being acceptable.
It doesn't have to work on all or even a majority of customers; as long as it works on just a small percentage, it's still going to increase profits. As long as it's not alienating the customers, it's all good from the corporate point of view, and attempting to guess what customers will be comfortable with helps.
I'm just surprised that it's Kingston Fried Chicken that starts with this. I would have thought that MD would have been first, given how much they already attempt to push products (to the point that I've driven out of the line in anger and went to BK instead).
I work as a programmer for KFC. Here is the source code for the recommendation algorithm:
String
getRecommendation(Customer customer)
{
switch (getCustomerPreference(customer.facialImage)) {
default:
return "Chicken";
}
}