Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386907915> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W4386907915 abstract "Annotating 3D LiDAR point clouds for perception tasks including 3D object detection and LiDAR semantic segmentation is notoriously time-and-energy-consuming. To alleviate the burden from labeling, it is promising to perform large-scale pre-training and fine-tune the pre-trained backbone on different downstream datasets as well as tasks. In this paper, we propose SPOT, namely Scalable Pre-training via Occupancy prediction for learning Transferable 3D representations, and demonstrate its effectiveness on various public datasets with different downstream tasks under the label-efficiency setting. Our contributions are threefold: (1) Occupancy prediction is shown to be promising for learning general representations, which is demonstrated by extensive experiments on plenty of datasets and tasks. (2) SPOT uses beam re-sampling technique for point cloud augmentation and applies class-balancing strategies to overcome the domain gap brought by various LiDAR sensors and annotation strategies in different datasets. (3) Scalable pre-training is observed, that is, the downstream performance across all the experiments gets better with more pre-training data. We believe that our findings can facilitate understanding of LiDAR point clouds and pave the way for future exploration in LiDAR pre-training. Codes and models will be released." @default.
- W4386907915 created "2023-09-21" @default.
- W4386907915 creator A5000349453 @default.
- W4386907915 creator A5008886708 @default.
- W4386907915 creator A5011877804 @default.
- W4386907915 creator A5012459112 @default.
- W4386907915 creator A5056327596 @default.
- W4386907915 creator A5058342038 @default.
- W4386907915 creator A5062676523 @default.
- W4386907915 creator A5072861783 @default.
- W4386907915 creator A5078348692 @default.
- W4386907915 creator A5087158377 @default.
- W4386907915 date "2023-09-19" @default.
- W4386907915 modified "2023-09-27" @default.
- W4386907915 title "SPOT: Scalable 3D Pre-training via Occupancy Prediction for Autonomous Driving" @default.
- W4386907915 doi "https://doi.org/10.48550/arxiv.2309.10527" @default.
- W4386907915 hasPublicationYear "2023" @default.
- W4386907915 type Work @default.
- W4386907915 citedByCount "0" @default.
- W4386907915 crossrefType "posted-content" @default.
- W4386907915 hasAuthorship W4386907915A5000349453 @default.
- W4386907915 hasAuthorship W4386907915A5008886708 @default.
- W4386907915 hasAuthorship W4386907915A5011877804 @default.
- W4386907915 hasAuthorship W4386907915A5012459112 @default.
- W4386907915 hasAuthorship W4386907915A5056327596 @default.
- W4386907915 hasAuthorship W4386907915A5058342038 @default.
- W4386907915 hasAuthorship W4386907915A5062676523 @default.
- W4386907915 hasAuthorship W4386907915A5072861783 @default.
- W4386907915 hasAuthorship W4386907915A5078348692 @default.
- W4386907915 hasAuthorship W4386907915A5087158377 @default.
- W4386907915 hasBestOaLocation W43869079151 @default.
- W4386907915 hasConcept C119857082 @default.
- W4386907915 hasConcept C121332964 @default.
- W4386907915 hasConcept C124101348 @default.
- W4386907915 hasConcept C127313418 @default.
- W4386907915 hasConcept C131979681 @default.
- W4386907915 hasConcept C134306372 @default.
- W4386907915 hasConcept C153294291 @default.
- W4386907915 hasConcept C154945302 @default.
- W4386907915 hasConcept C160331591 @default.
- W4386907915 hasConcept C162324750 @default.
- W4386907915 hasConcept C18903297 @default.
- W4386907915 hasConcept C21547014 @default.
- W4386907915 hasConcept C2776151529 @default.
- W4386907915 hasConcept C2776207758 @default.
- W4386907915 hasConcept C2777211547 @default.
- W4386907915 hasConcept C33923547 @default.
- W4386907915 hasConcept C36503486 @default.
- W4386907915 hasConcept C41008148 @default.
- W4386907915 hasConcept C48044578 @default.
- W4386907915 hasConcept C51399673 @default.
- W4386907915 hasConcept C62649853 @default.
- W4386907915 hasConcept C77088390 @default.
- W4386907915 hasConcept C86803240 @default.
- W4386907915 hasConcept C89600930 @default.
- W4386907915 hasConceptScore W4386907915C119857082 @default.
- W4386907915 hasConceptScore W4386907915C121332964 @default.
- W4386907915 hasConceptScore W4386907915C124101348 @default.
- W4386907915 hasConceptScore W4386907915C127313418 @default.
- W4386907915 hasConceptScore W4386907915C131979681 @default.
- W4386907915 hasConceptScore W4386907915C134306372 @default.
- W4386907915 hasConceptScore W4386907915C153294291 @default.
- W4386907915 hasConceptScore W4386907915C154945302 @default.
- W4386907915 hasConceptScore W4386907915C160331591 @default.
- W4386907915 hasConceptScore W4386907915C162324750 @default.
- W4386907915 hasConceptScore W4386907915C18903297 @default.
- W4386907915 hasConceptScore W4386907915C21547014 @default.
- W4386907915 hasConceptScore W4386907915C2776151529 @default.
- W4386907915 hasConceptScore W4386907915C2776207758 @default.
- W4386907915 hasConceptScore W4386907915C2777211547 @default.
- W4386907915 hasConceptScore W4386907915C33923547 @default.
- W4386907915 hasConceptScore W4386907915C36503486 @default.
- W4386907915 hasConceptScore W4386907915C41008148 @default.
- W4386907915 hasConceptScore W4386907915C48044578 @default.
- W4386907915 hasConceptScore W4386907915C51399673 @default.
- W4386907915 hasConceptScore W4386907915C62649853 @default.
- W4386907915 hasConceptScore W4386907915C77088390 @default.
- W4386907915 hasConceptScore W4386907915C86803240 @default.
- W4386907915 hasConceptScore W4386907915C89600930 @default.
- W4386907915 hasLocation W43869079151 @default.
- W4386907915 hasOpenAccess W4386907915 @default.
- W4386907915 hasPrimaryLocation W43869079151 @default.
- W4386907915 hasRelatedWork W2954561862 @default.
- W4386907915 hasRelatedWork W2976989770 @default.
- W4386907915 hasRelatedWork W3009777820 @default.
- W4386907915 hasRelatedWork W3091155238 @default.
- W4386907915 hasRelatedWork W3214186465 @default.
- W4386907915 hasRelatedWork W4226195147 @default.
- W4386907915 hasRelatedWork W4310264062 @default.
- W4386907915 hasRelatedWork W4312828305 @default.
- W4386907915 hasRelatedWork W4313048795 @default.
- W4386907915 hasRelatedWork W4385801324 @default.
- W4386907915 isParatext "false" @default.
- W4386907915 isRetracted "false" @default.
- W4386907915 workType "article" @default.