In this paper, we develop the continuous time dynamic topic model (cDTM). The cDTM is a dynamic topic model that uses Brownian motion to model the latent topics through a sequenti...
Abstract This paper focuses on human behavior recognition where the main problem is to bridge the semantic gap between the analogue observations of the real world and the symbolic ...
—We present a discriminative part-based approach for human action recognition from video sequences using motion features. Our model is based on the recently proposed hidden condi...
Markov random field (MRF) theory provides a basis for modeling contextual constraints in visual processing and interpretation. It enables us to develop optimal vision algorithms sy...
In this paper, we describe a statistical method to detect highlights in a baseball game video. The input video is first segmented into scene shots, within which the camera motion ...