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

Fine-grained 3D object detection is a core ability for agents to understandtheir 3D environment and interact with surrounding objects. However, currentmethods and benchmarks mainly focus on relatively large stuff. 3D objectdetectors still struggle on small objects due to weak geometric information.With in-depth study, we find increasing the spatial resolution of the featuremaps significantly boosts the performance of 3D small object detection. Andmore interestingly, though the computational overhead increases dramaticallywith resolution, the growth mainly comes from the upsampling operation of thedecoder. Inspired by this, we present a high-resolution multi-level detectorwith dynamic spatial pruning named DSPDet3D, which detects objects from largeto small by iterative upsampling and meanwhile prunes the spatialrepresentation of the scene at regions where there is no smaller object to bedetected in higher resolution. We organize two benchmarks on ScanNet andTO-SCENE dataset to evaluate the ability of fine-grained 3D object detection,where our DSPDet3D improves the detection performance of small objects to a newlevel while achieving leading inference speed compared with existing 3D objectdetection methods. Moreover, DSPDet3D trained with only ScanNet rooms cangeneralize well to scenes in larger scale. It takes less than 2s for DSPDet3Dto directly process a whole house or building consisting of dozens of roomswhile detecting out almost all objects, ranging from bottles to beds, on asingle RTX 3090 GPU. Project page: https://xuxw98.github.io/DSPDet3D/.

DSPDet3D: Dynamic Spatial Pruning for 3D Small Object Detection