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

Monocular image-based 3D perception has become an active research area inrecent years owing to its applications in autonomous driving. Approaches tomonocular 3D perception including detection and tracking, however, often yieldinferior performance when compared to LiDAR-based techniques. Throughsystematic analysis, we identified that per-object depth estimation accuracy isa major factor bounding the performance. Motivated by this observation, wepropose a multi-level fusion method that combines different representations(RGB and pseudo-LiDAR) and temporal information across multiple frames forobjects (tracklets) to enhance per-object depth estimation. Our proposed fusionmethod achieves the state-of-the-art performance of per-object depth estimationon the Waymo Open Dataset, the KITTI detection dataset, and the KITTI MOTdataset. We further demonstrate that by simply replacing estimated depth withfusion-enhanced depth, we can achieve significant improvements in monocular 3Dperception tasks, including detection and tracking.

Depth Estimation Matters Most: Improving Per-Object Depth Estimation for Monocular 3D Detection and Tracking