SLAP Tracker

A Multi-Frame Optical Flow Spot Tracker
Jizhou Li |Christopher Gilliam | Thierry Blu

Key points:

  • We propose a novel, multi-frame, tracker that exploits the stationary motion. More precisely, we first estimate the stationary motion and then use it to guide the spot tracker.

  • We obtain the stationary motion by adapting a recent optical flow algorithm [1] that relates one image to another locally using an all-pass filter. We perform this operation over all the image frames simultaneously and estimate a single, stationary optical flow.

  • We compare the proposed tracker with two existing techniques and show that our approach is more robust to high noise and varying structure.

J. Li, C. Gilliam and T. Blu, “A Multi-Frame Optical Flow Spot Tracker,” in Proc. IEEE Int. Conf. Image Process. (ICIP 2015), Quebec City, Canada.

1. Stationary Motion Estimation

Deterministic Dynamic Models

Stochastic Dynamic Models

2. Tracking Results

Deterministic Dynamic Models

Stochastic Dynamic Models

3. Real Microscopy Images [2]

The Estimated stationary motion field and tracking results


  • [1] C. Gilliam and T. Blu, “Local all-pass filters for optical flow estimation.,” in Proc. IEEE Int. Conf. Acoust Speech Signal Process. ((ICASSP 2015)), Brisbane, Australia, April 2015

  • [2] L. Serbus, B.-J. Cha, W. Theurkauf, and W. Saxton, “Dynein and the actin cytoskeleton control kinesin-driven cytoplasmic streaming in drosophila oocytes,” Development, vol. 132, no. 16, pp. 3743–3752, 2005.