Applying Multimodal Sensing to Indoor Localization

Indoor localization has been a tantalizing problem in mobile computing, and despite significant research, there is no solution yet in the mainstream. In this talk, I will discuss the landscape of indoor localization. I will also talk about my own research, UnLoc, which breaks away from pure RF based localization (e.g., cellular, Wi-Fi) and shows the benefits of leveraging smartphone sensors (accelerometers, gyroscopes, magnetometers, etc.) into the solution framework. I will describe additional solutions, VideoLoc, where feeds from surveillance cameras can be leveraged for highly precise localization and customer interaction, without compromising privacy of individual users. I will end with how some of our core techniques are not specific to localization and can be extended to other applications such as augmented reality.