Gregorij Kurillo

TL;DR: A controlled multimodal dataset consisting of temporally synchronized and geometrically calibrated data from an optical motion capture system, multi-baseline stereo cameras from multiple views, depth sensors, accelerometers and microphones, provides researchers an inclusive testbed to develop and benchmark new algorithms across multiple modalities under known capture conditions in various research domains.

TL;DR: A new representation of human actions called Sequence of the Most Informative Joints (SMIJ), which is extremely easy to interpret and performs better than several state-of-the-art algorithms for the task of human action recognition.

TL;DR: This paper compares the Kinect pose estimation (skeletonization) with more established techniques for pose estimation from motion capture data, examining the accuracy of joint localization and robustness of pose estimation with respect to the orientation and occlusions.

TL;DR: This work proposes a hierarchy of dynamic medial axis structures at several spatio-temporal scales that can be modeled using a set of Linear Dynamical Systems (LDSs), and proposes novel discriminative metrics for comparing these sets of LDSs for the task of human activity recognition.

TL;DR: Overall Kinect 2 has more robust and more accurate tracking of human pose as compared to Kinect 1, and is compared with an optical motion capture system.