BEV-IO: Enhancing Bird's-Eye-View 3D Detection with Instance Occupancy

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Zhang, Zaibin, Zhang, Yuanhang, Wang, Lijun, Wang, Yifan, Lu, Huchuan
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909068038242304
author Zhang, Zaibin
Zhang, Yuanhang
Wang, Lijun
Wang, Yifan
Lu, Huchuan
author_facet Zhang, Zaibin
Zhang, Yuanhang
Wang, Lijun
Wang, Yifan
Lu, Huchuan
contents A popular approach for constructing bird's-eye-view (BEV) representation in 3D detection is to lift 2D image features onto the viewing frustum space based on explicitly predicted depth distribution. However, depth distribution can only characterize the 3D geometry of visible object surfaces but fails to capture their internal space and overall geometric structure, leading to sparse and unsatisfactory 3D representations. To mitigate this issue, we present BEV-IO, a new 3D detection paradigm to enhance BEV representation with instance occupancy information. At the core of our method is the newly-designed instance occupancy prediction (IOP) module, which aims to infer point-level occupancy status for each instance in the frustum space. To ensure training efficiency while maintaining representational flexibility, it is trained using the combination of both explicit and implicit supervision. With the predicted occupancy, we further design a geometry-aware feature propagation mechanism (GFP), which performs self-attention based on occupancy distribution along each ray in frustum and is able to enforce instance-level feature consistency. By integrating the IOP module with GFP mechanism, our BEV-IO detector is able to render highly informative 3D scene structures with more comprehensive BEV representations. Experimental results demonstrate that BEV-IO can outperform state-of-the-art methods while only adding a negligible increase in parameters (0.2%) and computational overhead (0.24%in GFLOPs).
format Preprint
id arxiv_https___arxiv_org_abs_2305_16829
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle BEV-IO: Enhancing Bird's-Eye-View 3D Detection with Instance Occupancy
Zhang, Zaibin
Zhang, Yuanhang
Wang, Lijun
Wang, Yifan
Lu, Huchuan
Computer Vision and Pattern Recognition
A popular approach for constructing bird's-eye-view (BEV) representation in 3D detection is to lift 2D image features onto the viewing frustum space based on explicitly predicted depth distribution. However, depth distribution can only characterize the 3D geometry of visible object surfaces but fails to capture their internal space and overall geometric structure, leading to sparse and unsatisfactory 3D representations. To mitigate this issue, we present BEV-IO, a new 3D detection paradigm to enhance BEV representation with instance occupancy information. At the core of our method is the newly-designed instance occupancy prediction (IOP) module, which aims to infer point-level occupancy status for each instance in the frustum space. To ensure training efficiency while maintaining representational flexibility, it is trained using the combination of both explicit and implicit supervision. With the predicted occupancy, we further design a geometry-aware feature propagation mechanism (GFP), which performs self-attention based on occupancy distribution along each ray in frustum and is able to enforce instance-level feature consistency. By integrating the IOP module with GFP mechanism, our BEV-IO detector is able to render highly informative 3D scene structures with more comprehensive BEV representations. Experimental results demonstrate that BEV-IO can outperform state-of-the-art methods while only adding a negligible increase in parameters (0.2%) and computational overhead (0.24%in GFLOPs).
title BEV-IO: Enhancing Bird's-Eye-View 3D Detection with Instance Occupancy
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2305.16829