PVP-SSD: Point-Voxel Fusion with Partitioned Point Cloud Sampling for Anchor-Free Single-Stage Small 3D Object Detection

Simple item page
dc.contributor.authorWu, Xinlinen_US
dc.contributor.authorTian, Yibinen_US
dc.contributor.authorPan, Yinen_US
dc.contributor.authorZhang, Zhiyuanen_US
dc.contributor.authorWu, Xuesongen_US
dc.contributor.authorWang, Ruishengen_US
dc.contributor.authorZeng, Zhien_US
dc.contributor.editorChen, Renjieen_US
dc.contributor.editorRitschel, Tobiasen_US
dc.contributor.editorWhiting, Emilyen_US
dc.date.accessioned2024-10-13T18:03:25Z
dc.date.available2024-10-13T18:03:25Z
dc.date.issued2024
dc.description.abstractSingle-stage object detection from 3D point clouds in autonomous driving faces significant challenges, particularly in accurately detecting small objects. To address this issue, we propose a novel method called Point-Voxel dual-branch feature extraction with Partitioned point cloud sampling for anchor-free Single-Stage Detection of 3D objects (PVP-SSD). The network comprises two branches: a point branch and a voxel branch. In the point branch, a partitioned point cloud sampling strategy leverages axial features to divide the point cloud. Then, it assigns different sampling weights to various segments to enhance the sampling accuracy. Additionally, a local feature enhancement module explicitly calculates the correlation between key points and query points, improving the extraction of local features. In the voxel branch, we use 3D sparse convolution to extract instance structural features efficiently. The point-voxel dual-branch fusion dynamically integrates instance features extracted from both branches using a self-attention mechanism, which contains not only the category information of the detected object but also the spatial dimensions and heading angle. Consequently, PVP-SSD achieves a certain balance between preserving detailed information and maintaining structural integrity. Experimental results on the KITTI and ONCE datasets demonstrate that PVP-SSD excels in multi-category small 3D object detection.en_US
dc.description.sectionheadersPoint Cloud Processing and Analysis I
dc.description.seriesinformationPacific Graphics Conference Papers and Posters
dc.identifier.doi10.2312/pg.20241279
dc.identifier.isbn978-3-03868-250-9
dc.identifier.pages12 pages
dc.identifier.urihttps://doi.org/10.2312/pg.20241279
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/pg20241279
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectKeywords: 3D object detection, autonomous driving, partitioned point cloud, self-attention mechanism CCS Concepts: Computing methodologies −→Computer vision; Model development and analysis
dc.subject3D object detection
dc.subjectautonomous driving
dc.subjectpartitioned point cloud
dc.subjectself
dc.subjectattention mechanism CCS Concepts
dc.subjectComputing methodologies −→Computer vision
dc.subjectModel development and analysis
dc.titlePVP-SSD: Point-Voxel Fusion with Partitioned Point Cloud Sampling for Anchor-Free Single-Stage Small 3D Object Detectionen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
pg20241279.pdf
Size:
20.73 MB
Format:
Adobe Portable Document Format
Download
Collections
PG2024 Conference Papers and Posters