English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multi-sensor fusion is essential for an accurate and reliable autonomousdriving system. Recent approaches are based on point-level fusion: augmentingthe LiDAR point cloud with camera features. However, the camera-to-LiDARprojection throws away the semantic density of camera features, hindering theeffectiveness of such methods, especially for semantic-oriented tasks (such as3D scene segmentation). In this paper, we break this deeply-rooted conventionwith BEVFusion, an efficient and generic multi-task multi-sensor fusionframework. It unifies multi-modal features in the shared bird's-eye view (BEV)representation space, which nicely preserves both geometric and semanticinformation. To achieve this, we diagnose and lift key efficiency bottlenecksin the view transformation with optimized BEV pooling, reducing latency by morethan 40x. BEVFusion is fundamentally task-agnostic and seamlessly supportsdifferent 3D perception tasks with almost no architectural changes. Itestablishes the new state of the art on nuScenes, achieving 1.3% higher mAP andNDS on 3D object detection and 13.6% higher mIoU on BEV map segmentation, with1.9x lower computation cost. Code to reproduce our results is available athttps://github.com/mit-han-lab/bevfusion.

BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation