Visual computing for XR

computer-graphics, HCI, AI, visualization computer vision

AR visualisation

raw data (filtering) focus data (mapping) geometric data (rendering) image for AR need to combine raw data with environment

data: geometric data (mesh), camera image, registration data.

challenges with visual coherence: - occlusion - shadows - height in visual field (depth) - relative size - cluttering - geometric alignment, and occlusion

contextual data. with 3d models we can occlude.

some system do online real time 3d reconstruction for occlusion. and also identify and label objects in the scene

ghosting map to show pipes underground. (edges, saliency, texture details, etc)

2.5D is stereo -> 2d image with depth