How to Dominate ARCore 1.x : Build 9 Augmented Reality Apps

How well do you know ARCore?

Today we are going to learn all about how to implement a point cloud and plane detection using ARCore within Unity! We’ll create a simple app that replaces the floor with another material such as wood and even grass.

Today we are going build upon our last video point and plane detection and just add some hit testing functionality. So if you remember in the HelloAR app, when we touched the screen, we instantiated and anchored Andy the Android man onto the plane. This uses a concept called Ray casting to see where the touch point coordinates on our screen coincides with the plane that we detected. Now to make this tutorial a bit interesting, let’s add some physics and mesh colliders on our plane as well as on our virtual objects so that they interact with each other.

Today will be a short tutorial on how read pixel intensity values from the cameras image to get your average light estimation value. And also how to use these values to show or  hide your 3D object. In this case we have a scary monster with glowing red eyes who creeps closer and closer to your camera every time you switch off your light switch

Measure distances in Augmented Reality, using ARCore in Unity SDK

Learn how to control virtual 3D obects in AR. In this lesson we learn to create a dragon and control it. 

Afraid of crashing your drone? In this tutorial I’ll be show how you can take a drone flying experience into Augmented Reality with ARCore in Unity. So taking apps in Augmented Reality is really quite simple once you get the hang of ARCore. I will show you, that you just need one simple unity package and the rest of the functionality can be handle by the function of the Gameobject. What I mean exactly is that you can develop the game functionality in Unity and then just add in this Unity package to instantiate your app in Augmented Reality. Lets take a look at how easy this is.

Create a Solar System in Augmented Reality using Googles ARCore in Unity. Really fun to play and learn in AR.

Transport yourself into another world using ARCore to create an Augmented Reality Portal.

In this tutorial I’ll be showing you how to implement your own holograms in augmented reality. Now the holographic models are 4D volumetric sequences that are captured using volumetric capture technology. And one of the main companies specializing in this is 4D views. Now why would you want to have a 4D model or a 4d volumetric sequence in your app or tool? Well for a number of reasons. Firstly, instead of manually animating a 3d character to perform a complex sequence, you can perform the same sequence using real people in a volumetric capture studio. This results in a much more realistic experience of the sequence. Like for example, to animate a shaolin monk to perform these complex movements would be quite challenging to do as well time consuming, however getting the real deal to do these moves would be way simpler. Now this is all provided you have access to a volumetric capture system near you. At the moment, these studios are scarce, but just like any other technology, there could be studios like these popping around just about everywhere or they may become even cheap enough for you to buy your very own 4d volumetric capture system.

Now its not only 4dviews that are innovating in this space. Intel is investing millions if not billions into volumetric capture technology. Like if you watch this video on filmmaking in volumetric video, you capture once and thereafter you are able to manipulate any camera angle that you desire just as if you were dragging around your camera just like you would in unity within a 3d scene. This is not only great in film making but also in sports with intels Free-D technology. This amazing tech allows the viewer to view any angle of a sporting event in real time and really immerse yourself in the action. Other applications on the top of my head is an AR business card, where you can project a 4d hologram of yourself demonstrating or discussing a product such as if you are a yoga instructor. You can perhaps show your clients a pose and allow them to view any angle to help them better understand how to perform that pose. The world is your oyster here if you can apply this amazing technology to tourism, business, arts, engineering to name a few.

Okay so let me show you how to create your own 4d volumetric app using a demo model from 4d Views.

Hey guys and welcome back. In this tutorial I’ll be showing you how to implement vertical plane detection using the latest ARCore 1.2 SDK. Now in this lecture, I will show you how you can hang a virtual picture frame on your wall which is a really simple task. So to make this lecture a bit more interesting, I will show you how to integrate basic gesture control scripts into your main ARSceneController script. So we will cover Tap to spawn, which you have been doing though out this course, and then pinch to zoom, to make the picture larger or smaller as well as to rotate the image using a single touch. Now there are better and more robust gesture control scripts online and in the unity asset store. However, the purpose here is to show you how to implement your own scripts to for gesture touch control so that you have a better fundamental understanding of how commercial assets work.Now before we get started, remember that you will require ARCore 1.2 or later. This is because vertical plane detection was only recently launch in this version at Google IO developer conference 2018. You can use assets such as Lean touch or finger touch from the unity asset store.

ARCore 1.2 Features Augmented Images

Concluding the ARCore Course.

Watch the very first live interview podcast of Ritesh Kanjee, founder and CEO of Augmented Startups. He talks about his journey to this point, some of his courses, future of AR and a few fun facts!

Want to learn how to build your OWN Spiderman EDITH Glasses AI prototype? Watch this tutorial series to see how I prototype my own glasses with an Intelligent AI.

Project EDITH Glasses DIY - Phase 1 : AI Face Detection (Spider-Man Far From Home) Hey welcome back, So I am really cool and excited to start this new series which will be to try and reproduce the tech from Spiderman Far from home, the EDITH glasses.

Not sure if you guys have seen the movie, but spoiler alert … essentially Tony Stark hands down his Super Smart AI Glasses called EDITH which stands for Even Dead, Im The Hero … Classic tony. So these glasses give Peter Parker access to all of Stark Tech and also has smart Augmented Reality capability.

This series will try and push the boundaries of existing technology and see how many features we can fit in this project. So just some acknowledgements to some really cool people who have worked on EDITH Glasses are The Hacksmith, JLaservideo.

So before we dive into our approach to this technology lets check what has already been done. We are going to take Augmented Reality to the next level!

Lets see how to make smart AI glasses in this video.

Hey guys and welcome back. So, in the last lecture we made our attempt at face detection in Phase 1 of the EDITH glasses DIY prototype. So essentially, we used Unity with OpenCV to perform AI face detection using a ResNet CNN. In this video we are going to build on top this prototype and upgrade it with facial recognition.

As you can see from the demo after training the model, we are able to recognize the faces of 3 people, where we then instantiate a graphical UI for each persons Identity. This will allow us to get close to the Spiderman E.D.I.T.H. glasses functionality that we saw in the movie. We are going to take Augmented Reality to the next level!

Hey guys and welcome back. So in tutorial, I'm going to show you how to add AI object detection functionality to our EDITH Glasses prototype. So, what is cool about this phase is that we are able to detect multiple objects simultaneously and in real-time. This makes it really easy to integrate the detections with a Heads-Up-Display (HUD) GUI animation.

If you missed the previous phases of Project EDITH. We prototyped facial recognition and face detection to be able to identify people similar to how it was done in Spiderman Far from Home.

In this video, Phase 3 we are going to continue object detection using a MobileNet AI model using OpenCV for Unity. There was an option for using Yolo but it was more process intensive and we have to keep processing power in the back of our minds, should we wish to port all this to a mobile platform like android.

E.D.I.T.H Glasses - Phase 4 Tutorial : Build your own JARVIS and EDITH AI (Spider-Man Far From Home) Project EDITH Hey guys and welcome back, I hope you enjoyed that little demo of Project EDITH Phase 4! So this demo shows of an AI Assistant capability, whereby we are able to demo a design and development application using our EDITH Glasses.

We were able to call on JARVIS or EDITH to assist us in the design of the EDITH smart glasses. Now from the demo, there are a few cinematic enhancements that you should be aware off. First off the quality of the Text to Speech (TTS) engine using IBM Watson is different.

So both these voices are from the IBM Watson TTS Engine. The better sounding one is produced using an Enhanced Deep Neural Network (a DNN) for Text to Speech synthesis whereas the old one uses traditional methods.

Wrapping up the ARCore Course