Cargando…

WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving

Three-dimensional object detection is crucial for autonomous driving to understand the driving environment. Since the pooling operation causes information loss in the standard CNN, we designed a wavelet-multiresolution-analysis-based 3D object detection network without a pooling operation. Additiona...

Descripción completa

Detalles Bibliográficos
Autores principales: Alaba, Simegnew Yihunie, Ball, John E.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9505771/
https://www.ncbi.nlm.nih.gov/pubmed/36146359
http://dx.doi.org/10.3390/s22187010
_version_ 1784796556202868736
author Alaba, Simegnew Yihunie
Ball, John E.
author_facet Alaba, Simegnew Yihunie
Ball, John E.
author_sort Alaba, Simegnew Yihunie
collection PubMed
description Three-dimensional object detection is crucial for autonomous driving to understand the driving environment. Since the pooling operation causes information loss in the standard CNN, we designed a wavelet-multiresolution-analysis-based 3D object detection network without a pooling operation. Additionally, instead of using a single filter like the standard convolution, we used the lower-frequency and higher-frequency coefficients as a filter. These filters capture more relevant parts than a single filter, enlarging the receptive field. The model comprises a discrete wavelet transform (DWT) and an inverse wavelet transform (IWT) with skip connections to encourage feature reuse for contrasting and expanding layers. The IWT enriches the feature representation by fully recovering the lost details during the downsampling operation. Element-wise summation was used for the skip connections to decrease the computational burden. We trained the model for the Haar and Daubechies (Db4) wavelets. The two-level wavelet decomposition result shows that we can build a lightweight model without losing significant performance. The experimental results on KITTI’s BEV and 3D evaluation benchmark show that our model outperforms the PointPillars-based model by up to 14% while reducing the number of trainable parameters.
format Online
Article
Text
id pubmed-9505771
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-95057712022-09-24 WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving Alaba, Simegnew Yihunie Ball, John E. Sensors (Basel) Article Three-dimensional object detection is crucial for autonomous driving to understand the driving environment. Since the pooling operation causes information loss in the standard CNN, we designed a wavelet-multiresolution-analysis-based 3D object detection network without a pooling operation. Additionally, instead of using a single filter like the standard convolution, we used the lower-frequency and higher-frequency coefficients as a filter. These filters capture more relevant parts than a single filter, enlarging the receptive field. The model comprises a discrete wavelet transform (DWT) and an inverse wavelet transform (IWT) with skip connections to encourage feature reuse for contrasting and expanding layers. The IWT enriches the feature representation by fully recovering the lost details during the downsampling operation. Element-wise summation was used for the skip connections to decrease the computational burden. We trained the model for the Haar and Daubechies (Db4) wavelets. The two-level wavelet decomposition result shows that we can build a lightweight model without losing significant performance. The experimental results on KITTI’s BEV and 3D evaluation benchmark show that our model outperforms the PointPillars-based model by up to 14% while reducing the number of trainable parameters. MDPI 2022-09-16 /pmc/articles/PMC9505771/ /pubmed/36146359 http://dx.doi.org/10.3390/s22187010 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Alaba, Simegnew Yihunie
Ball, John E.
WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving
title WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving
title_full WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving
title_fullStr WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving
title_full_unstemmed WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving
title_short WCNN3D: Wavelet Convolutional Neural Network-Based 3D Object Detection for Autonomous Driving
title_sort wcnn3d: wavelet convolutional neural network-based 3d object detection for autonomous driving
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9505771/
https://www.ncbi.nlm.nih.gov/pubmed/36146359
http://dx.doi.org/10.3390/s22187010
work_keys_str_mv AT alabasimegnewyihunie wcnn3dwaveletconvolutionalneuralnetworkbased3dobjectdetectionforautonomousdriving
AT balljohne wcnn3dwaveletconvolutionalneuralnetworkbased3dobjectdetectionforautonomousdriving