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

3D pedestrian detection is a challenging task in automated driving because pedestrians are relatively small, frequently occluded and easily confused with narrow vertical objects. LiDAR and camera are two commonly used sensor modalities for this task, which should provide complementary information. U...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2020-09
Hauptverfasser: Juncong Fei, Chen, Wenbo, Heidenreich, Philipp, Wirges, Sascha, Stiller, Christoph
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Juncong Fei
Chen, Wenbo
Heidenreich, Philipp
Wirges, Sascha
Stiller, Christoph
description 3D pedestrian detection is a challenging task in automated driving because pedestrians are relatively small, frequently occluded and easily confused with narrow vertical objects. LiDAR and camera are two commonly used sensor modalities for this task, which should provide complementary information. Unexpectedly, LiDAR-only detection methods tend to outperform multisensor fusion methods in public benchmarks. Recently, PointPainting has been presented to eliminate this performance drop by effectively fusing the output of a semantic segmentation network instead of the raw image information. In this paper, we propose a generalization of PointPainting to be able to apply fusion at different levels. After the semantic augmentation of the point cloud, we encode raw point data in pillars to get geometric features and semantic point data in voxels to get semantic features and fuse them in an effective way. Experimental results on the KITTI test set show that SemanticVoxels achieves state-of-the-art performance in both 3D and bird's eye view pedestrian detection benchmarks. In particular, our approach demonstrates its strength in detecting challenging pedestrian cases and outperforms current state-of-the-art approaches.
doi_str_mv 10.48550/arxiv.2009.12276
format Article
fullrecord <record><control><sourceid>proquest_arxiv</sourceid><recordid>TN_cdi_arxiv_primary_2009_12276</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2446782187</sourcerecordid><originalsourceid>FETCH-LOGICAL-a527-f29986a9474ea32c42c7c325bfb9bf576b57eb4a85ec577c53c150ff901106703</originalsourceid><addsrcrecordid>eNo1kEFLAzEQhYMgWGp_gCcDnrcmk2Sz6620VoWCxRavSzZNSsp2U5Ndqf_etNXTDDNvHm8-hO4oGfNCCPKowtF9j4GQckwBZH6FBsAYzQoOcINGMe4IIZBLEIINkF-ZvWo7pz_90TTxCa_MV2_SQDV43kfnW2x9wGyGl2ZjYhecavHMdEZ3p11StFu8cLPJB15613Z42vh-g1W7wf_Oqdnuk6U6Xdyia6uaaEZ_dYjW8-f19DVbvL-8TSeLTAmQmYWyLHJVcsmNYqA5aKkZiNrWZW2FzGshTc1VIYwWUmrBNBXE2pJQSnJJ2BDdX2zPNKpDcHsVfqoTlepMJSkeLopD8Onj2FU734c2ZaqA81wWQAvJfgFZwGUc</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2446782187</pqid></control><display><type>article</type><title>SemanticVoxels: Sequential Fusion for 3D Pedestrian Detection using LiDAR Point Cloud and Semantic Segmentation</title><source>arXiv.org</source><source>Free E- Journals</source><creator>Juncong Fei ; Chen, Wenbo ; Heidenreich, Philipp ; Wirges, Sascha ; Stiller, Christoph</creator><creatorcontrib>Juncong Fei ; Chen, Wenbo ; Heidenreich, Philipp ; Wirges, Sascha ; Stiller, Christoph</creatorcontrib><description>3D pedestrian detection is a challenging task in automated driving because pedestrians are relatively small, frequently occluded and easily confused with narrow vertical objects. LiDAR and camera are two commonly used sensor modalities for this task, which should provide complementary information. Unexpectedly, LiDAR-only detection methods tend to outperform multisensor fusion methods in public benchmarks. Recently, PointPainting has been presented to eliminate this performance drop by effectively fusing the output of a semantic segmentation network instead of the raw image information. In this paper, we propose a generalization of PointPainting to be able to apply fusion at different levels. After the semantic augmentation of the point cloud, we encode raw point data in pillars to get geometric features and semantic point data in voxels to get semantic features and fuse them in an effective way. Experimental results on the KITTI test set show that SemanticVoxels achieves state-of-the-art performance in both 3D and bird's eye view pedestrian detection benchmarks. In particular, our approach demonstrates its strength in detecting challenging pedestrian cases and outperforms current state-of-the-art approaches.</description><identifier>EISSN: 2331-8422</identifier><identifier>DOI: 10.48550/arxiv.2009.12276</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Benchmarks ; Computer Science - Computer Vision and Pattern Recognition ; Computer Science - Robotics ; Image segmentation ; Lidar ; Multisensor fusion ; Object recognition ; Pedestrians ; Semantic segmentation ; Semantics ; Three dimensional models</subject><ispartof>arXiv.org, 2020-09</ispartof><rights>2020. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,780,881,27902</link.rule.ids><backlink>$$Uhttps://doi.org/10.1109/MFI49285.2020.9235240$$DView published paper (Access to full text may be restricted)$$Hfree_for_read</backlink><backlink>$$Uhttps://doi.org/10.48550/arXiv.2009.12276$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Juncong Fei</creatorcontrib><creatorcontrib>Chen, Wenbo</creatorcontrib><creatorcontrib>Heidenreich, Philipp</creatorcontrib><creatorcontrib>Wirges, Sascha</creatorcontrib><creatorcontrib>Stiller, Christoph</creatorcontrib><title>SemanticVoxels: Sequential Fusion for 3D Pedestrian Detection using LiDAR Point Cloud and Semantic Segmentation</title><title>arXiv.org</title><description>3D pedestrian detection is a challenging task in automated driving because pedestrians are relatively small, frequently occluded and easily confused with narrow vertical objects. LiDAR and camera are two commonly used sensor modalities for this task, which should provide complementary information. Unexpectedly, LiDAR-only detection methods tend to outperform multisensor fusion methods in public benchmarks. Recently, PointPainting has been presented to eliminate this performance drop by effectively fusing the output of a semantic segmentation network instead of the raw image information. In this paper, we propose a generalization of PointPainting to be able to apply fusion at different levels. After the semantic augmentation of the point cloud, we encode raw point data in pillars to get geometric features and semantic point data in voxels to get semantic features and fuse them in an effective way. Experimental results on the KITTI test set show that SemanticVoxels achieves state-of-the-art performance in both 3D and bird's eye view pedestrian detection benchmarks. In particular, our approach demonstrates its strength in detecting challenging pedestrian cases and outperforms current state-of-the-art approaches.</description><subject>Benchmarks</subject><subject>Computer Science - Computer Vision and Pattern Recognition</subject><subject>Computer Science - Robotics</subject><subject>Image segmentation</subject><subject>Lidar</subject><subject>Multisensor fusion</subject><subject>Object recognition</subject><subject>Pedestrians</subject><subject>Semantic segmentation</subject><subject>Semantics</subject><subject>Three dimensional models</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><sourceid>GOX</sourceid><recordid>eNo1kEFLAzEQhYMgWGp_gCcDnrcmk2Sz6620VoWCxRavSzZNSsp2U5Ndqf_etNXTDDNvHm8-hO4oGfNCCPKowtF9j4GQckwBZH6FBsAYzQoOcINGMe4IIZBLEIINkF-ZvWo7pz_90TTxCa_MV2_SQDV43kfnW2x9wGyGl2ZjYhecavHMdEZ3p11StFu8cLPJB15613Z42vh-g1W7wf_Oqdnuk6U6Xdyia6uaaEZ_dYjW8-f19DVbvL-8TSeLTAmQmYWyLHJVcsmNYqA5aKkZiNrWZW2FzGshTc1VIYwWUmrBNBXE2pJQSnJJ2BDdX2zPNKpDcHsVfqoTlepMJSkeLopD8Onj2FU734c2ZaqA81wWQAvJfgFZwGUc</recordid><startdate>20200925</startdate><enddate>20200925</enddate><creator>Juncong Fei</creator><creator>Chen, Wenbo</creator><creator>Heidenreich, Philipp</creator><creator>Wirges, Sascha</creator><creator>Stiller, Christoph</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PHGZM</scope><scope>PHGZT</scope><scope>PIMPY</scope><scope>PKEHL</scope><scope>PQEST</scope><scope>PQGLB</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20200925</creationdate><title>SemanticVoxels: Sequential Fusion for 3D Pedestrian Detection using LiDAR Point Cloud and Semantic Segmentation</title><author>Juncong Fei ; Chen, Wenbo ; Heidenreich, Philipp ; Wirges, Sascha ; Stiller, Christoph</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a527-f29986a9474ea32c42c7c325bfb9bf576b57eb4a85ec577c53c150ff901106703</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Benchmarks</topic><topic>Computer Science - Computer Vision and Pattern Recognition</topic><topic>Computer Science - Robotics</topic><topic>Image segmentation</topic><topic>Lidar</topic><topic>Multisensor fusion</topic><topic>Object recognition</topic><topic>Pedestrians</topic><topic>Semantic segmentation</topic><topic>Semantics</topic><topic>Three dimensional models</topic><toplevel>online_resources</toplevel><creatorcontrib>Juncong Fei</creatorcontrib><creatorcontrib>Chen, Wenbo</creatorcontrib><creatorcontrib>Heidenreich, Philipp</creatorcontrib><creatorcontrib>Wirges, Sascha</creatorcontrib><creatorcontrib>Stiller, Christoph</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection (ProQuest)</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>ProQuest Central (New)</collection><collection>ProQuest One Academic (New)</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Middle East (New)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Applied &amp; Life Sciences</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection><collection>arXiv Computer Science</collection><collection>arXiv.org</collection><jtitle>arXiv.org</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Juncong Fei</au><au>Chen, Wenbo</au><au>Heidenreich, Philipp</au><au>Wirges, Sascha</au><au>Stiller, Christoph</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>SemanticVoxels: Sequential Fusion for 3D Pedestrian Detection using LiDAR Point Cloud and Semantic Segmentation</atitle><jtitle>arXiv.org</jtitle><date>2020-09-25</date><risdate>2020</risdate><eissn>2331-8422</eissn><abstract>3D pedestrian detection is a challenging task in automated driving because pedestrians are relatively small, frequently occluded and easily confused with narrow vertical objects. LiDAR and camera are two commonly used sensor modalities for this task, which should provide complementary information. Unexpectedly, LiDAR-only detection methods tend to outperform multisensor fusion methods in public benchmarks. Recently, PointPainting has been presented to eliminate this performance drop by effectively fusing the output of a semantic segmentation network instead of the raw image information. In this paper, we propose a generalization of PointPainting to be able to apply fusion at different levels. After the semantic augmentation of the point cloud, we encode raw point data in pillars to get geometric features and semantic point data in voxels to get semantic features and fuse them in an effective way. Experimental results on the KITTI test set show that SemanticVoxels achieves state-of-the-art performance in both 3D and bird's eye view pedestrian detection benchmarks. In particular, our approach demonstrates its strength in detecting challenging pedestrian cases and outperforms current state-of-the-art approaches.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><doi>10.48550/arxiv.2009.12276</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2020-09
issn 2331-8422
language eng
recordid cdi_arxiv_primary_2009_12276
source arXiv.org; Free E- Journals
subjects Benchmarks
Computer Science - Computer Vision and Pattern Recognition
Computer Science - Robotics
Image segmentation
Lidar
Multisensor fusion
Object recognition
Pedestrians
Semantic segmentation
Semantics
Three dimensional models
title SemanticVoxels: Sequential Fusion for 3D Pedestrian Detection using LiDAR Point Cloud and Semantic Segmentation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-18T23%3A09%3A12IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_arxiv&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=SemanticVoxels:%20Sequential%20Fusion%20for%203D%20Pedestrian%20Detection%20using%20LiDAR%20Point%20Cloud%20and%20Semantic%20Segmentation&rft.jtitle=arXiv.org&rft.au=Juncong%20Fei&rft.date=2020-09-25&rft.eissn=2331-8422&rft_id=info:doi/10.48550/arxiv.2009.12276&rft_dat=%3Cproquest_arxiv%3E2446782187%3C/proquest_arxiv%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2446782187&rft_id=info:pmid/&rfr_iscdi=true