For those who can’t think of what purpose AR can be used .. — The idea that AR is sensing and a concrete example Cut out paper with AR
Here are some ideas on what purpose AR can be used and specific examples.
In a concrete example, we register an image and cut out that part from real paper.
We don’t know how to use AR
AR seems to be fun, but I don’t know what kind of application it should be used for.
The key is how to interact with the real thing
Understanding the underlying reality is key to augment reality.
AR is sensing
To augment reality, we need to understand “reality”.
In addition to displaying content, AR is the “sensing”.
The placement of AR content is just one of the part to augment the
reality, and where the content is placed in the reality and what reality is augmented by it may shape the purpose of AR.
Then , first of all, the sensing part of how to recognize the reality becomes important .
ARKit started with world tracking and face tracking, and recently such as understanding objects with lidar, the key is to grasp the reality with sensors,
Let’s think from reality
Let’s start with the reality that should be understood, not the virtual content.
What kind of reality is around you?
How can it be sensed and understood by a computer?
Finally, how can we augment that reality to produce interesting and useful effects?
example
Register the image and cut out that part from the actual paper.
There are various images around us.
Magazine covers, posters, advertisements.
Many people’s faces are there.
Fortunately, thanks to machine learning, we can recognize images and detect faces.
This is the sensing part.
Let’s give AR the reality of a face and augment it.
Augmentation isn’t just about adding.
Eliminating (reducing) it is also an augmentation.
Let’s cut out the grasped face and lose it.
First is the face grasping phase.
Vision detects faces from the camera frame and creates a face-only image.
Register the created face image as ARReferenceImage.
This allows ARKit to recognize facial images, create image anchors and place content.
Recognize the face registered with ARKit, place a box with the face image pasted on it, and move it to make it look like a clipping.
Let’s start from around
What’s around you that your computer can recognize?
How does a computer recognize those things?
And how can you augment it?
🐣
I’m a freelance engineer.
Work consultation
Please feel free to contact us with a brief development description.
rockyshikoku@gmail.com
I am making an app that uses Core ML and ARKit.
We send machine learning / AR related information.
