Studying intracellular dynamics is of major importance for understanding healthy life at the molecular level and for developing drugs to target disease processes. One of the key technologies to enable this research is the automated tracking and motion analysis of subcellular objects in microscopy image sequences. Contrary to common frame-by-frame tracking methods, two alternative approaches have been proposed recently, based on either Bayesian estimation or space-time segmentation, which better exploit the available spatiotemporal information. In this paper, we propose to combine the power of both approaches, and develop a new probabilistic method to segment the traces of the moving objects in kymograph representations of the image data. It is based on variable-rate particle filtering and uses multiscale trend analysis for estimation of the relevant kinematic parameters using the extracted traces. Experiments on realistic synthetically generated images as well as on real biological image data demonstrate the improved potential of the new method for the analysis of microtubule dynamics in vitro.