Sub-frame Appearance and 6D Pose Estimation of Fast Moving Objects

TL;DR

This paper introduces TbD-3D, a method for precise 6D pose and appearance estimation of fast-moving objects using sub-frame localization, enabling super-resolution and shape accuracy.

Contribution

It presents a novel reconstruction algorithm for sub-frame object localization and appearance estimation, and introduces a new dataset for fast-moving object tracking.

Findings

Accurate 6D pose estimation at sub-frame resolution.

Effective shape and appearance reconstruction of fast objects.

A new dataset with high-speed recordings and ground-truth annotations.

Abstract

We propose a novel method that tracks fast moving objects, mainly non-uniform spherical, in full 6 degrees of freedom, estimating simultaneously their 3D motion trajectory, 3D pose and object appearance changes with a time step that is a fraction of the video frame exposure time. The sub-frame object localization and appearance estimation allows realistic temporal super-resolution and precise shape estimation. The method, called TbD-3D (Tracking by Deblatting in 3D) relies on a novel reconstruction algorithm which solves a piece-wise deblurring and matting problem. The 3D rotation is estimated by minimizing the reprojection error. As a second contribution, we present a new challenging dataset with fast moving objects that change their appearance and distance to the camera. High speed camera recordings with zero lag between frame exposures were used to generate videos with different frame rates annotated with ground-truth trajectory and pose.