# Dense 3D Visual Mapping via Semantic Simplification

## TL;DR

This paper introduces semantic-aware simplification methods for dense 3D visual mapping, reducing noise and redundancy by selectively simplifying regions based on semantic segmentation, and fusing point clouds with Delaunay Triangulation.

## Contribution

It proposes four novel semantic-based point cloud simplification techniques that preserve important details and boundaries, improving the quality of dense 3D models.

## Key findings

- Semantic simplification reduces noise in point clouds.
- Preserves edges at class boundaries for better model accuracy.
- Simplifies models while maintaining essential details.

## Abstract

Dense 3D visual mapping estimates as many as possible pixel depths, for each image. This results in very dense point clouds that often contain redundant and noisy information, especially for surfaces that are roughly planar, for instance, the ground or the walls in the scene. In this paper we leverage on semantic image segmentation to discriminate which regions of the scene require simplification and which should be kept at high level of details. We propose four different point cloud simplification methods which decimate the perceived point cloud by relying on class-specific local and global statistics still maintaining more points in the proximity of class boundaries to preserve the infra-class edges and discontinuities. 3D dense model is obtained by fusing the point clouds in a 3D Delaunay Triangulation to deal with variable point cloud density. In the experimental evaluation we have shown that, by leveraging on semantics, it is possible to simplify the model and diminish the noise affecting the point clouds.