Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Event-ECC: Asynchronous Tracking of Events with Continuous Optimization

Maria Zafeiri, Georgios Evangelidis, Emmanouil Psarakis

TL;DR

An event-based tracker is presented that adopts the Enhanced Correlation Coefficient (ECC) criterion and proposes a tracking algorithm that computes a 2D motion warp per single event, called event-ECC (eECC).

Abstract

In this paper, an event-based tracker is presented. Inspired by recent advances in asynchronous processing of individual events, we develop a direct matching scheme that aligns spatial distributions of events at different times. More specifically, we adopt the Enhanced Correlation Coefficient (ECC) criterion and propose a tracking algorithm that computes a 2D motion warp per single event, called event-ECC (eECC). The complete tracking of a feature along time is cast as a \emph{single} iterative continuous optimization problem, whereby every single iteration is executed per event. The computational burden of event-wise processing is alleviated through a lightweight version that benefits from incremental processing and updating scheme. We test the proposed algorithm on publicly available datasets and we report improvements in tracking accuracy and feature age over state-of-the-art event-based asynchronous trackers.

Event-ECC: Asynchronous Tracking of Events with Continuous Optimization

TL;DR

An event-based tracker is presented that adopts the Enhanced Correlation Coefficient (ECC) criterion and proposes a tracking algorithm that computes a 2D motion warp per single event, called event-ECC (eECC).

Abstract

In this paper, an event-based tracker is presented. Inspired by recent advances in asynchronous processing of individual events, we develop a direct matching scheme that aligns spatial distributions of events at different times. More specifically, we adopt the Enhanced Correlation Coefficient (ECC) criterion and propose a tracking algorithm that computes a 2D motion warp per single event, called event-ECC (eECC). The complete tracking of a feature along time is cast as a \emph{single} iterative continuous optimization problem, whereby every single iteration is executed per event. The computational burden of event-wise processing is alleviated through a lightweight version that benefits from incremental processing and updating scheme. We test the proposed algorithm on publicly available datasets and we report improvements in tracking accuracy and feature age over state-of-the-art event-based asynchronous trackers.
Paper Structure (12 sections, 23 equations, 5 figures, 1 table, 1 algorithm)

This paper contains 12 sections, 23 equations, 5 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Problem Formulation
  4. Preliminaries
  5. Initialization Phase
  6. Tracking the Feature $F$
  7. Event tracking using ECC Criterion
  8. Model and Template Windows
  9. The Proposed Optimization Criterion
  10. Lightweight event-ECC
  11. Experiments
  12. Conclusions

Figures (5)

  • Figure 1: (a): Tracking trajectories starting from different seeds (white points) in "shapes" scene with 6DoF motion. Unlike HasteCorrelation$^\ast$, eECC generates smooth and continuous trajectories. (b): Final template from tracking the same feature for same age. eECC generates sharper templates (e.g. horizontal edges) because of more accurate tracking and motion compensation. To quantify this difference, we apply the same low threhold into the templates and measure the percentage of the area that generates events. Ideally, such binary masks should resemble the edge-map of the top-left image of Fig. \ref{['fig:mesh1']} and this percentage should be less than $10\%$.
  • Figure 2: Reprojection error and the cumulative distribution of outliers for the "wallposter" scene
  • Figure 3: Reprojection error and the cumulative distribution of outliers for the "boxes" scene
  • Figure 4: Reprojection error and the cumulative distribution of outliers for the "dynamic" scene
  • Figure 5: Reprojection error and the cumulative distribution of outliers for HDR recordings