Home

I. Introduction

In the previous version of Underworlds, the perception pipeline was separated from higher level reasoning, after experimenting with the original version it became clear that higher level reasoning need strong interaction with the perception layer to benefit from then and to not loose information in the abstraction.

Underworlds 3 is a framework to design perception pipelines with high level reasoning capabilities like physical reasoning, perspective taking, human beliefs estimation, social signal processing or language models for human-robot interaction.

Camera pinhole model

In underworlds, the pinhole camera model is very important as it is used to project into a plane the 3D scene. If you are not familiar with this concept read this. The scene nodes, represent at the same time the 3D entity in the scene and the 2d bounding box represented by the camera that observe it. One scene node can have multiple representation depending how many camera is looking at it.

II. Base types

Vector

Uwds3 have it's own vector library that implement for each one a stable version of it by using a linear kalman model. This allow to tradeoff accuracy and stability, two important parameters when dealing with physical simulation.

Shape

BoundingBox

Detection

FacialLandmarks

TemporalSituation

SceneNode

III. Reasoning