Vivo Unveils 3D Sensing Technology With 10 Times as Many Data Points as Apple's Face ID

Yea, but does it “just work”?

My FaceID on the iPhone X didn't just work

which it says has 10 times the accuracy of the Face ID authentication system in Apple's iPhone X.

Vivo says its Time of Flight (TOF) system uses 300,000 data points to map the user's face in three dimensions, compared to the 30,000 points of infrared light used in Apple's facial mapping technology.

Point of clarification: 10 times the data points does not automatically mean 10 times the accuracy, at least from a usability standpoint. I can draw a straight line, for example, with two points or 50 points and be equally accurate. Of course faces are more complicated, but I’m pointing out that more data points aren’t necessarily more accurate. In fact, if the logic behind the data points isn’t smart, such a setup could theoretically actually be less accurate by interpreting subtle changes in facial expression, or a scar, or a shadow, as a different face.

My FaceID on the iPhone X didn't just work

How so? Mine is pretty flawless.

Which I understood, I just meant that by extrapolating your statement, Touch ID is a disaster for someone like my dad.

I don't think faceid or touchid is a failure but I think the term It just works, is an inaccurate over used meme that no longer describes apple products.

This is just pure dishonest claim by Vivo. (if the article is completely accurate) How can the number of data points be comparable if the underlying technology is completely different. According to this article Vivo is using Time of Flight which is very bad to track anything but motion (like Microsoft 2nd gen Kinect). Apple didn't use Time of Flight for face recognition. They only use Time of Flight as the proximity sensor to trigger rest of the FaceID main components. The main technology Apple use for FaceID is Structured-light which is far more accurate in 3d space recognition than Time of Flight can ever hope to achieve.

To explain why Time of Flight is bad for face recognition no matter how many data point they add. It is due to physic especially speed of the light. Time of Flight rely on detecting extremely tiny difference in time of travel for the light to bounce back from the target. The speed of light is extremely fast, and for the tiny distance between your face and where you normal hold your phone the difference in distance between how far is point A and point B on your face from your phone is too small to tell for any modern CPU (The accuracy is limited by the clock speed. for any CPU it can only detect difference no smaller than its single clock cycle, and for the time between a single clock cycle for a 3Ghz CPU the light can travel almost 10cm during that time. It can barely tell that you had a nose) It might be good for application such as autonomous driving sensor or AR/VR in large space where resolution of few centimeters isn't really that much an issue when combine with other sensors, but unless this article is wrong about which technology they use I wouldn't trust it any more than the pattern lock on Android phone.

Quite simple it wouldn't work all of the time and I had to type in my code instead.

Well by that logic, Touch ID was an utter failure. Especially if you've been a tradesman all of your life and your fingerprints have more in common with 100 grit sandpaper than skin.

This reminds me so much of the megapixel camera war. More pixels does not always mean a better picture.

The Apple logo does not always mean a better product.

Yea, but does it “just work”?

How so? Mine is pretty flawless.

Quite simple it wouldn't work all of the time and I had to type in my code instead.

How so? Mine is pretty flawless.

Mine also, it just works.