Resolving motion correspondence for densely moving points

Abstract

AbstractÐThis paper studies the motion correspondence problem for which a diversity of qualitative and statistical solutions exist. We concentrate on qualitative modeling, especially in situations where assignment conflicts arise either because multiple features compete for one detected point or because multiple detected points fit a single feature point. We leave out the possibility of point track initiation and termination because that principally conflicts with allowing for temporary point occlusion. We introduce individual, combined, and global motion models and fit existing qualitative solutions in this framework. Additionally, we present a new efficient tracking algorithm that satisfies theseÐpossibly constrainedÐmodels in a greedy matching sense, including an effective way to handle detection errors and occlusion. The performance evaluation shows that the proposed algorithm outperforms existing greedy matching algorithms. Finally, we describe an extension to the tracker that enables automatic initialization of the point tracks. Several experiments show that the extended algorithm is efficient, hardly sensitive its few parameters, and qualitatively better than other algorithms, including the presumed optimal statistical multiple hypothesis tracker.

Analytical tools for single-molecule fluorescence imaging in cellulo

“…Heuristic methods can be used to identify putative tracks from qualitative descriptions, but suffer the disadvantage of relying critically on the accuracy in the detection stage and easily fail when ambiguities occur. 48,49 To overcome many of the shortcomings of existing tracking methods, my laboratory has developed a new automated multiparticle tracking algorithm based on minimal path optimization, similar to those used previously but extended in application to native cell membranes of living bacterial cells. After detecting candidate particles and linking image feature points frame-byframe, some segmented trajectories are obtained initially.…”

Detection and tracking of clathrin-coated pits in biological images

“…Furthermore, only completely matched trajectories in the experiment were considered as correct. Figure 4 represents the performances of IPAN [14], 1-to-1 admission method [7], and N-to-N admission method for variable particle density. In the 1-to-1 admission method, one particle only corresponds to one particle in the next frame; while in the N-to-N method, multiple particles can correspond to one particle in the next frame, and one particle can correspond to multiple particles in the next frame.…”

“…If one point of any pair in the match has a better correspondence choice, this pair is recognized as unstable. The matching set is considered as stable only if no unstable pair exists [7].…”

“…A multiscale wavelet products method has been successfully applied to extract particles with arbitrary shape and size from 730 Liu Z F, et al Chin Sci Bull March (2012) Vol.57 No.7 a single image with an inhomogeneous gray level distribution of both background and particles [1,6]. However, it cannot be used directly in segmenting image series with a fixed threshold because of the image intensity variance over the entire image series.Particle tracking is concerned with the tracking of multiple moving objects in an image sequence and reconstruction of all of the trajectories of these objects over the entire sequence [7]. Recently, several algorithms have been developed for tracking the motion of particles and cells [8][9][10][11][12].…”

“…Particle tracking is concerned with the tracking of multiple moving objects in an image sequence and reconstruction of all of the trajectories of these objects over the entire sequence [7]. Recently, several algorithms have been developed for tracking the motion of particles and cells [8][9][10][11][12].…”