2D-3D Geometric Fusion Network using Multi-Neighbourhood Graph Convolution for RGB-D Indoor Scene Classification

2020-09-23 · Albert Mosella-Montoro, Javier Ruiz-Hidalgo

Multi-modal fusion has been proved to help enhance the performance of scene classification tasks. This paper presents a 2D-3D Fusion stage that combines 3D Geometric Features with 2D Texture Features obtained by 2D Convolutional Neural Networks. To get a robust 3D Geometric embedding, a network that uses two novel layers is proposed. The first layer, Multi-Neighbourhood Graph Convolution, aims to learn a more robust geometric descriptor of the scene combining two different neighbourhoods: one in the Euclidean space and the other in the Feature space. The second proposed layer, Nearest Voxel Pooling, improves the performance of the well-known Voxel Pooling. Experimental results, using NYU-Depth-V2 and SUN RGB-D datasets, show that the proposed method outperforms the current state-of-the-art in RGB-D indoor scene classification task.

📄 PDF Abstract BibTeX arXiv:2009.11154

Code (0)

등록된 구현이 없습니다.

Tasks

General Classification Scene Classification

Methods 이 논문이 사용한 방법론

Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…

Simple Spectral Graph Convolution

2021-01-01 · ICLR 2021 1 · Hao Zhu, Piotr Koniusz

Graph Convolutional Networks (GCNs) have drawn significant attention and become promising methods for learning graph representations. The most GCNs suffer the performance loss when the depth of the model increases. Simi…

Clustering Node Classification Node Clustering Text Classification

Characters as Graphs: Recognizing Online Handwritten Chinese Characters via Spatial Graph Convolutional Network

2020-04-20 · Ji Gan, Weiqiang Wang, Ke Lu

Chinese is one of the most widely used languages in the world, yet online handwritten Chinese character recognition (OLHCCR) remains challenging. To recognize Chinese characters, one popular choice is to adopt the 2D con…

Time Series Time Series Analysis

Capturing Graphs with Hypo-Elliptic Diffusions

2022-05-27 · Csaba Toth, Darrick Lee, Celia Hacker, Harald Oberhauser

Convolutional layers within graph neural networks operate by aggregating information about local neighbourhood structures; one common way to encode such substructures is through random walks. The distribution of these ra…

Graph Attention Network for Node Regression on Random Geometric Graphs with Erdős--Rényi contamination

2026-01-30 · Somak Laha, Suqi Liu, Morgane Austern arxiv

Graph attention networks (GATs) are widely used and often appear robust to noise in node covariates and edges, yet rigorous statistical guarantees demonstrating a provable advantage of GATs over non-attention graph neura…

Geometric and Topological Deep Learning for Predicting Thermo-mechanical Performance in Cold Spray Deposition Process Modeling

2026-03-15 · Akshansh Mishra arxiv

This study presents a geometric deep learning framework for predicting cold spray particle impact responses using finite element simulation data. A parametric dataset was generated through automated Abaqus simulations sp…

Graph Neural Network