Improving Lidar-Based Semantic Segmentation of Top-View Grid Maps by Learning Features in Complementary Representations

03/02/2022
by   Frank Bieder, et al.
1

In this paper we introduce a novel way to predict semantic information from sparse, single-shot LiDAR measurements in the context of autonomous driving. In particular, we fuse learned features from complementary representations. The approach is aimed specifically at improving the semantic segmentation of top-view grid maps. Towards this goal the 3D LiDAR point cloud is projected onto two orthogonal 2D representations. For each representation a tailored deep learning architecture is developed to effectively extract semantic information which are fused by a superordinate deep neural network. The contribution of this work is threefold: (1) We examine different stages within the segmentation network for fusion. (2) We quantify the impact of embedding different features. (3) We use the findings of this survey to design a tailored deep neural network architecture leveraging respective advantages of different representations. Our method is evaluated using the SemanticKITTI dataset which provides a point-wise semantic annotation of more than 23.000 LiDAR measurements.

READ FULL TEXT

page 1

page 3

page 5

research
05/13/2020

Exploiting Multi-Layer Grid Maps for Surround-View Semantic Segmentation of Sparse LiDAR Data

In this paper, we consider the transformation of laser range measurement...
research
08/30/2019

LU-Net: An Efficient Network for 3D LiDAR Point Cloud Semantic Segmentation Based on End-to-End-Learned 3D Features and U-Net

We propose LU-Net -- for LiDAR U-Net, a new method for the semantic segm...
research
11/03/2022

Semantic 3D Grid Maps for Autonomous Driving

Maps play a key role in rapidly developing area of autonomous driving. W...
research
07/14/2023

LEST: Large-scale LiDAR Semantic Segmentation with Transformer

Large-scale LiDAR-based point cloud semantic segmentation is a critical ...
research
09/25/2020

SemanticVoxels: Sequential Fusion for 3D Pedestrian Detection using LiDAR Point Cloud and Semantic Segmentation

3D pedestrian detection is a challenging task in automated driving becau...
research
02/09/2022

A Multi-Task Recurrent Neural Network for End-to-End Dynamic Occupancy Grid Mapping

A common approach for modeling the environment of an autonomous vehicle ...
research
08/24/2021

Learning Spatiotemporal Occupancy Grid Maps for Lifelong Navigation in Dynamic Scenes

We present a novel method for generating, predicting, and using Spatiote...

Please sign up or login with your details

Forgot password? Click here to reset