v2 Chapter 765： Amazing success rate

Mike Connor reached out and held the microphone on his face, with a faint smile on his face: "Before the report officially starts, I want to explain to you what computer vision is."

"Some people may imagine that computer vision is very complicated, but it is very simple and can be understood literally."

"Computer vision is actually a science that studies how to make machines "see". Furthermore, it refers to the use of cameras and computers instead of human eyes to recognize, track, and measure targets to become a technology more suitable for human observation."

"For this we have done a lot of research, such as image classification, object detection, target tracking, language segmentation, and instance segmentation."

"These research results have promoted the vigorous development of computer vision, so today, I want to propose another topic, which is to apply computer vision to the field of intelligent robots! Let intelligent robots have vision like humans!"

"Just recently our computer vision development team at Imperial College did a time. We have already built the neural network on a machine, let him replace humans for digital image forensics, and identify those that have been modified through real tampering image training. Past pictures."

Quiet!

Deathly quiet, everyone is staring at Mike Connor. They all know that this is the first experiment of computer vision in the field of robotics.

If successful, it will set off a huge storm in the computer field, which may change the pattern of several industries!

"Guru~" Everyone swallowed and asked cautiously: "What's the result? Did the test succeed?"

"It's not a success, it's not a failure..." After thinking about it, Mike Connor said slowly.

Success means success, failure means failure. What does it mean if this is not a success or a failure?

Forgive them for reading less, they don't understand what Mike Connor means.

"Professor Connor, what do you mean?" After a long silence, someone in the crowd asked.

"Before the experiment, we selected 20 pictures and modified 10 of them to some extent, but the results of the experiment were not satisfactory."

"After 10 changes, the computer only found 7 pictures!" Mike Conner said blankly: "So, for this result, I can't think he succeeded, nor can I think he failed."

The overall test process is OK, but the test results are not satisfactory.

If computer vision cannot be 100% correct, Mike Connor will not think he is successful.

Because a 100% success rate is very important.

Just like his face payment system in the future application field, if the computer can't even recognize the face 100%, it is easy to be stolen.

This will not only bring huge economic losses to people, but also have a great impact on the discipline of computer vision.

In the first trial, it has reached a 70% success rate!

What about the second and third time?

To put it bluntly, this is already an extremely high success rate.

They really can't figure out why Mike Connor is still not satisfied!

"Professor Connor, a 70% success rate was achieved in the first experiment. This is already a remarkable achievement..."