Calculation corner based on three surfaces Introduction Last time you have learnt about ARKit basics. You can build simple app with horizontal and vertical surface detection. You also know more about movement and rotation in the 3D space. With prepared tools, you can calculate the corner of the room and place whatever you want there. …
Category: Advanced
Corner and surface detection in AR Part 2
Plane movement and rotation in 3D space Introduction Last time you have learnt about basics of ARKit and how to use it to detect surfaces. You should also know how to get input from the plugin and calculate angles between detected surfaces. That allows you to gather three perpendicular planes which is the first step …
Corner and surface detection in AR Part 1
Introduction AR technology is getting more and more popular these days. Two big companies have their own implementation of it. Google has ARCore supporting devices with Android and Apple has ARKit for iOS. These technologies require high computing power so only newer devices are supported. Both ARCore and ARKit have a feature to detect horizontal …
Artificial locomotion in Unity using Machine Learning Part 3
Training and results ML-Agents training One of the most essential parts of training is hyperparameter tuning. Effective training using RL requires to setup up hyperparameters properly. In ML-Agents all hyperparameters are specified in trainer_config.yaml file. Here you can find definitions for all hyperparameters. Below you can see the trainer_config.yaml file with hyperparameters config used to …
Artificial locomotion in Unity using Machine Learning Part 2
Implementation Previously, in part 1, we looked at the theory of both ML-agents deep learning algorithms and evolutionary computation. Now it’s time for us to implement the whole thing. First, let’s start by making a creature! Creature The creature we’re going to train will be designed in 2D environment, for the sake of simplicity. On …