Disney Research Pittsburgh

We present a hierarchical model for human activity recognition in entire multi-person scenes.

Automatic recovery of 3d pose of multiple interacting subjects from unconstrained monocular image sequence is a challenging and largely unaddressed problem.

We present a type of Temporal Restricted Boltzmann Machine that defines a probability distribution over an output sequence conditional on an input sequence.

This work reports the first experimental demonstration of CIT from a vertical magnetic dipole (VMD) in remote sensing and position tracking applications.

In this work, we show that the estimation errors are frequency dependent and that by combining multiple frequency signals, the overall RMS estimation error can be significantly reduced.

We present freeform interactive speakers for creating spatial sound experiences from a variety of surfaces. Traditional surround sound systems are widely used and consist of multiple electromagnetic speakers that create point sound sources within a space.

We present freeform interactive loudspeakers for creating spatial sound experiences from a variety of surfaces.

In this paper, a transducer to monitor the curvature of a tagged object for backscatter RFID sensors operating at 5:8GHz is presented.

In this paper we report on a magnetoquasistatic orientation sensor that uses the magnetoquasistatic coupling between an electrically small emitting loop (magnetic dipole) and seven vertical receiving loops located on a circle of radius 12.19 m to determine the orientation of an object.

This paper presents such a backscatter sensor system operating in the 5.8 GHz unlicensed frequency band that wirelessly senses displacement between one and five millimeters.