Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale Outdoor Scene

Sunghwan Yoo, Yeongjeong Jeong, Maryam Jameela, Gunho Sohn
1/30/2023
cs.CV

No official code found. Best implementation: Yacovitch/EyeNet (16 stars, last commit 2025-04-03).

Abstract

This paper proposes EyeNet, a novel semantic segmentation network for point clouds that addresses the critical yet often overlooked parameter of coverage area size. Inspired by human peripheral vision, EyeNet overcomes the limitations of conventional networks by introducing a simple but efficient multi-contour input and a parallel processing network with connection blocks between parallel streams. The proposed approach effectively addresses the challenges of dense point clouds, as demonstrated by our ablation studies and state-of-the-art performance on Large-Scale Outdoor datasets.

View on arXivView PDF

Code Implementations(1)

Yacovitch/EyeNetCommunity100%
161C, C++Jun 12, 20231 years agoNOASSERTION

Similar Papers

Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs

86% match1 repo8012017

Deep Learning-Based 3D Instance and Semantic Segmentation: A Review

86% match2024

Small but Mighty: Enhancing 3D Point Clouds Semantic Segmentation with U-Next Framework

85% match1 repo232023

RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds

85% match1 repo1.6k2019

RESSCAL3D: Resolution Scalable 3D Semantic Segmentation of Point Clouds

85% match2024

APNet: Urban-level Scene Segmentation of Aerial Images and Point Clouds

84% match2 repos2.0k2023

Cite this paper

@article{yoo2023human,
  title  = {Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale Outdoor Scene},
  author = {Sunghwan Yoo and Yeongjeong Jeong and Maryam Jameela and Gunho Sohn},
  year   = {2023},
  eprint = {2301.12972},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2301.12972v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2301.12972.svg)](https://codeofpaper.com/paper/2301.12972)

Discussion