A Neurobiologically Motivated Stochastic Method for Analysis of Human Activities in Video

In this paper, we develop a neurobiologicallymotivated statistical method for video analysis that simultaneously searches the combined motion and form space in a concerted and efficient manner using wellknown Markov chain Monte Carlo (MCMC) techniques. Specifically, we leverage upon an MCMC variant called the Hamiltonian Monte Carlo (HMC), which we extend to utilize data-based proposals rather than the blind proposals in a traditional HMC, thus creating the DataDriven HMC (DDHMC). We demonstrate the efficacy of our system on real-life video sequences.

• Type: Conference
• Year: 2010
• Where: ICPR
• Authors: Ricky Sethi, Amit Roy-Chowdhury
• Keywords: Computer Vision, Hamiltonian Monte Carlo, ICPR 2010, Markov Chain Monte Carlo, Monte Carlo