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

Annotating 3D LiDAR point clouds for perception tasks including 3D objectdetection and LiDAR semantic segmentation is notoriouslytime-and-energy-consuming. To alleviate the burden from labeling, it ispromising to perform large-scale pre-training and fine-tune the pre-trainedbackbone on different downstream datasets as well as tasks. In this paper, wepropose SPOT, namely Scalable Pre-training via Occupancy prediction forlearning Transferable 3D representations, and demonstrate its effectiveness onvarious public datasets with different downstream tasks under thelabel-efficiency setting. Our contributions are threefold: (1) Occupancyprediction is shown to be promising for learning general representations, whichis demonstrated by extensive experiments on plenty of datasets and tasks. (2)SPOT uses beam re-sampling technique for point cloud augmentation and appliesclass-balancing strategies to overcome the domain gap brought by various LiDARsensors and annotation strategies in different datasets. (3) Scalablepre-training is observed, that is, the downstream performance across all theexperiments gets better with more pre-training data. We believe that ourfindings can facilitate understanding of LiDAR point clouds and pave the wayfor future exploration in LiDAR pre-training. Codes and models will bereleased.

SPOT: Scalable 3D Pre-training via Occupancy Prediction for Autonomous Driving