# Visual Coin-Tracking: Tracking of Planar Double-Sided Objects

## TL;DR

This paper introduces the novel problem of tracking planar double-sided objects like coins in videos, presents a new benchmark dataset, and proposes a baseline CNN-based method demonstrating the challenge of this task.

## Contribution

It defines a new coin-tracking problem, provides a benchmark dataset with annotations, and develops a baseline segmentation and pose modeling approach.

## Key findings

- Sequences differ from standard tracking datasets
- Baseline method confirms the problem's difficulty
- Coin-tracking remains an open challenge

## Abstract

We introduce a new video analysis problem -- tracking of rigid planar objects in sequences where both their sides are visible. Such coin-like objects often rotate fast with respect to an arbitrary axis producing unique challenges, such as fast incident light and aspect ratio change and rotational motion blur. Despite being common, neither tracking sequences containing coin-like objects nor suitable algorithm have been published. As a second contribution, we present a novel coin-tracking benchmark containing 17 video sequences annotated with object segmentation masks. Experiments show that the sequences differ significantly from the ones encountered in standard tracking datasets. We propose a baseline coin-tracking method based on convolutional neural network segmentation and explicit pose modeling. Its performance confirms that coin-tracking is an open and challenging problem.