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

It is a long-term vision for Autonomous Driving (AD) community that theperception models can learn from a large-scale point cloud dataset, to obtainunified representations that can achieve promising results on different tasksor benchmarks. Previous works mainly focus on the self-supervised pre-trainingpipeline, meaning that they perform the pre-training and fine-tuning on thesame benchmark, which is difficult to attain the performance scalability andcross-dataset application for the pre-training checkpoint. In this paper, forthe first time, we are committed to building a large-scale pre-trainingpoint-cloud dataset with diverse data distribution, and meanwhile learninggeneralizable representations from such a diverse pre-training dataset. Weformulate the point-cloud pre-training task as a semi-supervised problem, whichleverages the few-shot labeled and massive unlabeled point-cloud data togenerate the unified backbone representations that can be directly applied tomany baseline models and benchmarks, decoupling the AD-related pre-trainingprocess and downstream fine-tuning task. During the period of backbonepre-training, by enhancing the scene- and instance-level distribution diversityand exploiting the backbone's ability to learn from unknown instances, weachieve significant performance gains on a series of downstream perceptionbenchmarks including Waymo, nuScenes, and KITTI, under different baselinemodels like PV-RCNN++, SECOND, CenterPoint.

AD-PT: Autonomous Driving Pre-Training with Large-scale Point Cloud Dataset