FBK home > INFORMATION TECHNOLOGY > Technologies of Vision > semantic image labelling

Technologies of Vision: semantic image labelling


go to marmota blog
go to marmota eagleeye

Visual Environment Monitoring: the Marmota project

Our goal is to understand what is being seen through the lens of a camera. Through the integration of diverse skills from many disciplines, ranging from cartographic projection, advanced 3D computer graphics and cutting-edge machine vision algorithms we hope to automatically align, generate and visualize geo-referenced meta-data (contacts).

wandering marmota mobile
3D from 2D Content Creation Social Sensor Network
3D from 2D
Content Creation
Social Sensor Network
By accurately aligning 2D photos with a 3D model of the real World we can bring an added dimension to your photos. No need for multiple images; just give Marmota a location and it does the rest By seeing inside photos we can auto-generate meta-data such as visible mountains or villages, estimate snow coverage and even segment foreground objects through occlusion reasoning A spatio-temporal overlay can be realised by amassing thousands of aligned images. From such a layer we can generate new, constantly updating and novel viewpoints

For further information please contact: