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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

3D object detection from multiple image views is a fundamental andchallenging task for visual scene understanding. Due to its low cost and highefficiency, multi-view 3D object detection has demonstrated promisingapplication prospects. However, accurately detecting objects throughperspective views in the 3D space is extremely difficult due to the lack ofdepth information. Recently, DETR3D introduces a novel 3D-2D query paradigm inaggregating multi-view images for 3D object detection and achievesstate-of-the-art performance. In this paper, with intensive pilot experiments,we quantify the objects located at different regions and find that the"truncated instances" (i.e., at the border regions of each image) are the mainbottleneck hindering the performance of DETR3D. Although it merges multiplefeatures from two adjacent views in the overlapping regions, DETR3D stillsuffers from insufficient feature aggregation, thus missing the chance to fullyboost the detection performance. In an effort to tackle the problem, we proposeGraph-DETR3D to automatically aggregate multi-view imagery information throughgraph structure learning (GSL). It constructs a dynamic 3D graph between eachobject query and 2D feature maps to enhance the object representations,especially at the border regions. Besides, Graph-DETR3D benefits from a noveldepth-invariant multi-scale training strategy, which maintains the visual depthconsistency by simultaneously scaling the image size and the object depth.Extensive experiments on the nuScenes dataset demonstrate the effectiveness andefficiency of our Graph-DETR3D. Notably, our best model achieves 49.5 NDS onthe nuScenes test leaderboard, achieving new state-of-the-art in comparisonwith various published image-view 3D object detectors.

Graph-DETR3D: Rethinking Overlapping Regions for Multi-View 3D Object Detection