Rendering and Tracking the Directional TSDF: Modeling Surface Orientation for Coherent Maps

TL;DR

This paper introduces methods for rendering depth and color maps from the Directional TSDF, enhancing map coherence and tracking performance in real-time RGB-D mapping for robotics.

Contribution

It presents rendering techniques for the Directional TSDF, enabling seamless integration into existing systems and improving scene reusability and color accuracy.

Findings

Enhanced map coherence with DTSDF

Improved re-usability of mapped scenes

Color integration improves surface color correctness

Abstract

Dense real-time tracking and mapping from RGB-D images is an important tool for many robotic applications, such as navigation or grasping. The recently presented Directional Truncated Signed Distance Function (DTSDF) is an augmentation of the regular TSDF and shows potential for more coherent maps and improved tracking performance. In this work, we present methods for rendering depth- and color maps from the DTSDF, making it a true drop-in replacement for the regular TSDF in established trackers. We evaluate and show, that our method increases re-usability of mapped scenes. Furthermore, we add color integration which notably improves color-correctness at adjacent surfaces.