# DenseBEV: Transforming BEV Grid Cells into 3D Objects

Source: [DenseBEV: Transforming BEV Grid Cells into 3D Objects](<https://www.sota2.com/research/paper/6944dc27423ea42265729cb8>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-20T05:00:10.481Z

Authors: Marius D\\"ahling, Sebastian Krebs, J. Marius Z\\"ollner

Published: 2025-12-18T17:59:22.000Z

## Abstract

In current research, Bird's-Eye-View (BEV)-based transformers are increasingly utilized for multi-camera 3D object detection. Traditional models often employ random queries as anchors, optimizing them successively. Recent advancements complement or replace these random queries with detections from auxiliary networks. We propose a more intuitive and efficient approach by using BEV feature cells directly as anchors. This end-to-end approach leverages the dense grid of BEV queries, considering each cell as a potential object for the final detection task. As a result, we introduce a novel two-stage anchor generation method specifically designed for multi-camera 3D object detection. To address the scaling issues of attention with a large number of queries, we apply BEV-based Non-Maximum Suppression, allowing gradients to flow only through non-suppressed objects. This ensures efficient training without the need for post-processing. By using BEV features from encoders such as BEVFormer directly as object queries, temporal BEV information is inherently embedded. Building on the temporal BEV information already embedded in our object queries, we introduce a hybrid temporal modeling approach by integrating prior detections to further enhance detection performance. Evaluating our method on the nuScenes dataset shows consistent and significant improvements in NDS and mAP over the baseline, even with sparser BEV grids and therefore fewer initial anchors. It is particularly effective for small objects, enhancing pedestrian detection with a 3.8% mAP increase on nuScenes and an 8% increase in LET-mAP on Waymo. Applying our method, named DenseBEV, to the challenging Waymo Open dataset yields state-of-the-art performance, achieving a LET-mAP of 60.7%, surpassing the previous best by 5.4%. Code is available at https://github.com/mdaehl/DenseBEV.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [3D Object Detection on nuScenes v1.0-trainval (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-nuscenes-v1-0-trainval-val.md>): 3D Object Detection; nuScenes v1.0-trainval (val); NDS: 54.9.

- [3D Object Detection on nuScenes (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-nuscenes-val.md>): 3D Object Detection; nuScenes (val); NDS: 54.9.

- [3D Object Detection on Waymo Open Dataset (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-waymo-open-dataset-val.md>): 3D Object Detection; Waymo Open Dataset (val); 3D APH Vehicle L2: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.16818>)

[PDF](<https://arxiv.org/pdf/2512.16818.pdf>)
