DeepAI AI Chat
Log In Sign Up

Efficient Surfel Fusion Using Normalised Information Distance

by   Louis Gallagher, et al.
Maynooth University

We present a new technique that achieves a significant reduction in the quantity of measurements required for a fusion based dense 3D mapping system to converge to an accurate, de-noised surface reconstruction. This is achieved through the use of a Normalised Information Distance metric, that computes the novelty of the information contained in each incoming frame with respect to the reconstruction, and avoids fusing those frames that exceed a redundancy threshold. This provides a principled approach for opitmising the trade-off between surface reconstruction accuracy and the computational cost of processing frames. The technique builds upon the ElasticFusion (EF) algorithm where we report results of the technique's scalability and the accuracy of the resultant maps by applying it to both the ICL-NUIM and TUM RGB-D datasets. These results demonstrate the capabilities of the approach in performing accurate surface reconstructions whilst utilising a fraction of the frames when compared to the original EF algorithm.


page 1

page 3

page 4


Collaborative Dense SLAM

In this paper, we present a new system for live collaborative dense surf...

High order low-bit Sigma-Delta quantization for fusion frames

We construct high order low-bit Sigma-Delta (ΣΔ) quantizers for the vect...

HRBF-Fusion: Accurate 3D reconstruction from RGB-D data using on-the-fly implicits

Reconstruction of high-fidelity 3D objects or scenes is a fundamental re...

Directional TSDF: Modeling Surface Orientation for Coherent Meshes

Real-time 3D reconstruction from RGB-D sensor data plays an important ro...

Log-GPIS-MOP: A Unified Representation for Mapping, Odometry and Planning

Whereas dedicated scene representations are required for each different ...