HV-BEV: Decoupling Horizontal and Vertical Feature Sampling for Multi-View 3D Object Detection Article Swipe

MAKE IMPACT

Di Wu , Feng Yang , Benlian Xu , Pan Liao , Wenhui Zhao , Dingwen Zhang · YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2412.18884

The application of vision-based multi-view environmental perception system has been increasingly recognized in autonomous driving technology, especially the BEV-based models. Current state-of-the-art solutions primarily encode image features from each camera view into the BEV space through explicit or implicit depth prediction. However, these methods often overlook the structured correlations among different parts of objects in 3D space and the fact that different categories of objects often occupy distinct local height ranges. For example, trucks appear at higher elevations, whereas traffic cones are near the ground. In this work, we propose a novel approach that decouples feature sampling in the \textbf{BEV} grid queries paradigm into \textbf{H}orizontal feature aggregation and \textbf{V}ertical adaptive height-aware reference point sampling (HV-BEV), aiming to improve both the aggregation of objects' complete information and awareness of diverse objects' height distribution. Specifically, a set of relevant neighboring points is dynamically constructed for each 3D reference point on the ground-aligned horizontal plane, enhancing the association of the same instance across different BEV grids, especially when the instance spans multiple image views around the vehicle. Additionally, instead of relying on uniform sampling within a fixed height range, we introduce a height-aware module that incorporates historical information, enabling the reference points to adaptively focus on the varying heights at which objects appear in different scenes. Extensive experiments validate the effectiveness of our proposed method, demonstrating its superior performance over the baseline across the nuScenes dataset. Moreover, our best-performing model achieves a remarkable 50.5\% mAP and 59.8\% NDS on the nuScenes testing set. The code is available at https://github.com/Uddd821/HV-BEV.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2412.18884

PDF

https://arxiv.org/pdf/2412.18884

OA Status

green

Related Works

10

OpenAlex ID

https://openalex.org/W4405900719

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4405900719

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2412.18884

Digital Object Identifier

Title

HV-BEV: Decoupling Horizontal and Vertical Feature Sampling for Multi-View 3D Object Detection

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2024

Year of publication

Publication date

2024-12-25

Full publication date if available

Authors

Di Wu, Feng Yang, Benlian Xu, Pan Liao, Wenhui Zhao, Dingwen Zhang

List of authors in order

Landing page

https://arxiv.org/abs/2412.18884

Publisher landing page

PDF URL

https://arxiv.org/pdf/2412.18884

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2412.18884

Direct OA link when available

Concepts

Decoupling (probability), Horizontal and vertical, Object (grammar), Feature (linguistics), Computer vision, Computer science, Artificial intelligence, Sampling (signal processing), Object detection, Pattern recognition (psychology), Geology, Geodesy, Engineering, Linguistics, Philosophy, Filter (signal processing), Control engineering

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload