Files

Abstract

This paper presents a probabilistic algorithm for segmenting text embedded in video based on Monte Carlo sampling. The algorithm approximates the posterior of segmentation thresholds of video text by a set of weighted samples, referred to as particles. The set of samples is initialized by applying a traditional segmentation algorithm on the first video frame and further refined by random sampling under a temporal Bayesian framework. Results on a database of 6944 images demonstrated the validity of the algorithm.

Details

Actions

Preview