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

Deploying 3D detectors in unfamiliar domains has been demonstrated to resultin a significant 70-90% drop in detection rate due to variations in lidar,geography, or weather from their training dataset. This domain gap leads tomissing detections for densely observed objects, misaligned confidence scores,and increased high-confidence false positives, rendering the detector highlyunreliable. To address this, we introduce MS3D++, a self-training framework formulti-source unsupervised domain adaptation in 3D object detection. MS3D++generates high-quality pseudo-labels, allowing 3D detectors to achieve highperformance on a range of lidar types, regardless of their density. Ourapproach effectively fuses predictions of an ensemble of multi-framepre-trained detectors from different source domains to improve domaingeneralization. We subsequently refine predictions temporally to ensuretemporal consistency in box localization and object classification.Furthermore, we present an in-depth study into the performance andidiosyncrasies of various 3D detector components in a cross-domain context,providing valuable insights for improved cross-domain detector ensembling.Experimental results on Waymo, nuScenes and Lyft demonstrate that detectorstrained with MS3D++ pseudo-labels achieve state-of-the-art performance,comparable to training with human-annotated labels in Bird's Eye View (BEV)evaluation for both low and high density lidar. Code is available athttps://github.com/darrenjkt/MS3D

MS3D++: Ensemble of Experts for Multi-Source Unsupervised Domain Adaptation in 3D Object Detection