Tracking multiple moving objects in images using Markov Chain Monte Carlo
This work addresses the challenge of accurate object tracking in noisy or low-visibility conditions, such as in microscopy, but it is incremental as it builds on existing Bayesian and MCMC frameworks.
The authors tackled the problem of tracking multiple moving objects in images by proposing a Bayesian MCMC algorithm that directly models the image generation process, avoiding information loss from point extraction. Their method showed improved tracking performance over conventional techniques, particularly for dim targets with overlapping illuminated regions in both synthetic and real microscopy data.
A new Bayesian state and parameter learning algorithm for multiple target tracking (MTT) models with image observations is proposed. Specifically, a Markov chain Monte Carlo algorithm is designed to sample from the posterior distribution of the unknown number of targets, their birth and death times, states and model parameters, which constitutes the complete solution to the tracking problem. The conventional approach is to pre-process the images to extract point observations and then perform tracking. We model the image generation process directly to avoid potential loss of information when extracting point observations. Numerical examples show that our algorithm has improved tracking performance over commonly used techniques, for both synthetic examples and real florescent microscopy data, especially in the case of dim targets with overlapping illuminated regions.