We show the first camera based (privacy-preserving) indoor mobile positioning system, CaPSuLe, which does not involve any communication (or data transfer) with any other device or the cloud. The algorithm only needs 78.9MB of memory and can localize a mobile device with 92.11% accuracy. Furthermore this is done in 1.92 seconds of on-device computation consuming 3.77 Joules of energy, as evaluated on Samsung Galaxy S4 platform. At the core, our solutions uses a smart implementation of hashing-based image matching algorithm which is more than 500x cheaper, both in energy and computation cost, over existing state-of-the-art matching techniques.

This significant reduction allows us to perform end-to-end computation locally on the mobile device. In contrast traditional approaches would consume 2100 Joules and takes more than 1000 seconds with a small accuracy increase of 0.89%.

The ability to run the complete algorithm on the mobile device eliminates the need for the cloud, making CaPSuLe a privacy-preserving localization algorithm by design as it does not require any communication

Related Papers

CaPSuLe: Camera Based Positioning System Using Learning .[pdf] [slides]

Yongshik Moon, Soonhyun Noh, Daedong Park, Chen Luo, Anshumali Shrivastava, Seongsoo Hong, Krishna Palem. (IEEE SOCC 2016)

Code Coming Soon...