# Deep 6-DOF Tracking

## TL;DR

This paper introduces a deep learning-based 6-DOF tracking method that achieves high accuracy and robustness to occlusions in real-time, evaluated on challenging RGBD datasets without hand-crafted features.

## Contribution

The paper presents a novel data-driven 6-DOF tracking approach that outperforms existing methods in accuracy and occlusion robustness while maintaining real-time performance.

## Key findings

- State-of-the-art accuracy on real-world datasets
- Enhanced robustness to occlusions
- Real-time tracking capability

## Abstract

We present a temporal 6-DOF tracking method which leverages deep learning to achieve state-of-the-art performance on challenging datasets of real world capture. Our method is both more accurate and more robust to occlusions than the existing best performing approaches while maintaining real-time performance. To assess its efficacy, we evaluate our approach on several challenging RGBD sequences of real objects in a variety of conditions. Notably, we systematically evaluate robustness to occlusions through a series of sequences where the object to be tracked is increasingly occluded. Finally, our approach is purely data-driven and does not require any hand-designed features: robust tracking is automatically learned from data.