AR Content Placement using 3D models and orthographic projections

 

Matt Ruge

Paul Plunkett

PROGRESS REPORT 2

ABSTRACT

We feel that Augmented Reality would be a much more useable and popular tool if resources didn’t require an investment of time, expertise, and money.  Our goal is to provide an interface that would allow simple augmented reality interfaces to be created in as little as an afternoon.

Estimated Progress from Last Proposal

 

Task

Completed

Setup cloud interface infrastructure, create basic Argon2 page for interfacing with cloud

yes

Load Models from Cloud

yes

(Spring break) Update Argon 2 page

See Below [1]

Update Desktop Version [2]

yes

 

[1] Due to debugging problems, work beyond initial loading of models in ARGON was moved to later.  A JAILBroken iPhone is capable of running WebGL in Safari.  This makes all development and demoing possible with Safari.  GUI model loading and THREE.js pages are all functioning in Safari.  Thus references to ARGON2 in original proposal can often be substituted by Safari, unless referencing the camera.

[2] The Desktop Version will load all ortho versions of objects and  can manually specify GPS location to load.

Additional Completed Work

 

Task

Completed

Obtain Models of GATECH campus

yes

3d object Selection, and such change camera orientation to be relative  to selected object.

yes

Three.js Collision detection

yes

 

Future Work (To be Completed by April. 22)

 

Snap based on Proximity, and allow for thresholds.

Provide visual feedback on what is ‘snappable’

UI and performance improvements

Comments are closed.