RadarCat

Abstract

In RadarCat we present a small, versatile radar-based system for material and object classification which enables new forms of everyday proximate interaction with digital devices. We demonstrate that we can train and classify different types of materials and objects which we can then recognize in real time. Based on established research designs, we report on the results of three studies, first with 26 materials (including complex composite objects), next with 16 transparent materials (with different thickness and varying dyes) and finally 10 body parts from 6 participants. Both leave one-out and 10-fold cross-validation demonstrate that our approach of classification of radar signals using random forest classifier is robust and accurate. We further demonstrate four working examples including a physical object dictionary, painting and photo editing application, body shortcuts and automatic refill based on RadarCat. We conclude with a discussion of our results, limitations and outline future directions.

Exploring Tangible Interactions with Radar Sensing

“…In the health domain, radar related sensing has been explored for monitoring glucose levels in blood samples non-invasively [37,52], breathing and heart rate monitoring [2], sleep sensing [44] and emotion tracking [71] in a smart home environment. In terms of object and material recognition, prior work in RadarCat [67,68] and Deskwave [33] have demonstrated the ability to disambiguate passive and unmodified objects and their materials by analyzing the characteristics of the radio frequencies reflected.…”

“…The use of radar is not new, indeed automotive, ground and weather radar are commonly employed at varying scales and with variable environmental and lighting conditions. The potential of radar sensing in HCI also has been demonstrated in detecting micro hand gestures [32,61] and in our prior work in recognizing objects and materials [68]. However, other facets of radar have remained underexplored in the context of HCI, especially in tangible interaction.…”

“…To realize our prototype, we employ the Google ATAP Project Soli radar sensor [20,32], extend the RadarCat [68] system and by building on the Token+Constraint [56] paradigm. We show that by adding a "guide case" that act as constraints, we can greatly expand the interaction capabilities and improve the object sensing accuracy.…”

Enabling Single-Handed Interaction in Mobile and Wearable Computing

“…In both RadarCat [9] and SpeCam [10], I started exploring new ways to use a device with simple one-handed gestures, building on the idea of placement-aware computing [2]. By using a miniature radar sensor (RadarCat) or the front-facing camera of smartphone (SpeCam), the device can then recognize the surface where it is placed or touched.…”

Tangible UI by object and material classification with radar

“…While TUI a empt to give digital information a physical and manipulable form. RadarCat [Yeo et al 2016] advances our ability to realize such visions with radar based sensing. It allows for new forms and styles of SA '17 Emerging Technologies, Bangkok, ailand 2017.…”