Abstract. This paper addresses the problem of probabilistically modeling 3D human motion for synthesis and tracking. Given the high dimensional nature of human motion, learning an ...
Background: Profile HMMs (hidden Markov models) provide effective methods for modeling the conserved regions of protein families. A limitation of the resulting domain models is th...
The ability to retrieve molecules based on structural similarity has use in many applications, from disease diagnosis and treatment to drug discovery and design. In this paper, we...
Keith Marsolo, Srinivasan Parthasarathy, Kotagiri ...
Sequence matching techniques are effective for comparing two videos. However, existing approaches suffer from demanding computational costs and thus are not scalable for large-sca...