Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object Detection

ABSTRACT Data augmentation plays an important role in boosting the performance of 3D models, while very few studies handle the 3D point cloud data with this technique. Global augmentation and cut‐paste are commonly used augmentation techniques for point clouds, where global augmentation is applied t...

Full description

Saved in:
Bibliographic Details
Main Authors: Huaijin Liu, Jixiang Du, Yong Zhang, Hongbo Zhang, Jiandian Zeng
Format: Article
Language:English
Published: Wiley 2025-06-01
Series:CAAI Transactions on Intelligence Technology
Subjects:
Online Access:https://doi.org/10.1049/cit2.70001
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849473011834421248
author Huaijin Liu
Jixiang Du
Yong Zhang
Hongbo Zhang
Jiandian Zeng
author_facet Huaijin Liu
Jixiang Du
Yong Zhang
Hongbo Zhang
Jiandian Zeng
author_sort Huaijin Liu
collection DOAJ
description ABSTRACT Data augmentation plays an important role in boosting the performance of 3D models, while very few studies handle the 3D point cloud data with this technique. Global augmentation and cut‐paste are commonly used augmentation techniques for point clouds, where global augmentation is applied to the entire point cloud of the scene, and cut‐paste samples objects from other frames into the current frame. Both types of data augmentation can improve performance, but the cut‐paste technique cannot effectively deal with the occlusion relationship between the foreground object and the background scene and the rationality of object sampling, which may be counterproductive and may hurt the overall performance. In addition, LiDAR is susceptible to signal loss, external occlusion, extreme weather and other factors, which can easily cause object shape changes, while global augmentation and cut‐paste cannot effectively enhance the robustness of the model. To this end, we propose Syn‐Aug, a synchronous data augmentation framework for LiDAR‐based 3D object detection. Specifically, we first propose a novel rendering‐based object augmentation technique (Ren‐Aug) to enrich training data while enhancing scene realism. Second, we propose a local augmentation technique (Local‐Aug) to generate local noise by rotating and scaling objects in the scene while avoiding collisions, which can improve generalisation performance. Finally, we make full use of the structural information of 3D labels to make the model more robust by randomly changing the geometry of objects in the training frames. We verify the proposed framework with four different types of 3D object detectors. Experimental results show that our proposed Syn‐Aug significantly improves the performance of various 3D object detectors in the KITTI and nuScenes datasets, proving the effectiveness and generality of Syn‐Aug. On KITTI, four different types of baseline models using Syn‐Aug improved mAP by 0.89%, 1.35%, 1.61% and 1.14% respectively. On nuScenes, four different types of baseline models using Syn‐Aug improved mAP by 14.93%, 10.42%, 8.47% and 6.81% respectively. The code is available at https://github.com/liuhuaijjin/Syn‐Aug.
format Article
id doaj-art-52661a99acee46ff92a60d751facfe8a
institution Kabale University
issn 2468-2322
language English
publishDate 2025-06-01
publisher Wiley
record_format Article
series CAAI Transactions on Intelligence Technology
spelling doaj-art-52661a99acee46ff92a60d751facfe8a2025-08-20T03:24:17ZengWileyCAAI Transactions on Intelligence Technology2468-23222025-06-0110391292810.1049/cit2.70001Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object DetectionHuaijin Liu0Jixiang Du1Yong Zhang2Hongbo Zhang3Jiandian Zeng4School of Artificial Intelligence Guangzhou Maritime University China and College of Mechanical Engineering and Automation Huaqiao University Xiamen ChinaCollege of Computer Science and Technology Huaqiao University Xiamen ChinaCollege of Mechanical Engineering and Automation Huaqiao University Xiamen ChinaCollege of Computer Science and Technology Huaqiao University Xiamen ChinaInstitute of Artificial Intelligence and Future Networks Beijing Normal University Zhuhai ChinaABSTRACT Data augmentation plays an important role in boosting the performance of 3D models, while very few studies handle the 3D point cloud data with this technique. Global augmentation and cut‐paste are commonly used augmentation techniques for point clouds, where global augmentation is applied to the entire point cloud of the scene, and cut‐paste samples objects from other frames into the current frame. Both types of data augmentation can improve performance, but the cut‐paste technique cannot effectively deal with the occlusion relationship between the foreground object and the background scene and the rationality of object sampling, which may be counterproductive and may hurt the overall performance. In addition, LiDAR is susceptible to signal loss, external occlusion, extreme weather and other factors, which can easily cause object shape changes, while global augmentation and cut‐paste cannot effectively enhance the robustness of the model. To this end, we propose Syn‐Aug, a synchronous data augmentation framework for LiDAR‐based 3D object detection. Specifically, we first propose a novel rendering‐based object augmentation technique (Ren‐Aug) to enrich training data while enhancing scene realism. Second, we propose a local augmentation technique (Local‐Aug) to generate local noise by rotating and scaling objects in the scene while avoiding collisions, which can improve generalisation performance. Finally, we make full use of the structural information of 3D labels to make the model more robust by randomly changing the geometry of objects in the training frames. We verify the proposed framework with four different types of 3D object detectors. Experimental results show that our proposed Syn‐Aug significantly improves the performance of various 3D object detectors in the KITTI and nuScenes datasets, proving the effectiveness and generality of Syn‐Aug. On KITTI, four different types of baseline models using Syn‐Aug improved mAP by 0.89%, 1.35%, 1.61% and 1.14% respectively. On nuScenes, four different types of baseline models using Syn‐Aug improved mAP by 14.93%, 10.42%, 8.47% and 6.81% respectively. The code is available at https://github.com/liuhuaijjin/Syn‐Aug.https://doi.org/10.1049/cit2.700013D object detectiondata augmentationdiversitygeneralizationpoint cloudrobustness
spellingShingle Huaijin Liu
Jixiang Du
Yong Zhang
Hongbo Zhang
Jiandian Zeng
Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object Detection
CAAI Transactions on Intelligence Technology
3D object detection
data augmentation
diversity
generalization
point cloud
robustness
title Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object Detection
title_full Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object Detection
title_fullStr Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object Detection
title_full_unstemmed Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object Detection
title_short Syn‐Aug: An Effective and General Synchronous Data Augmentation Framework for 3D Object Detection
title_sort syn aug an effective and general synchronous data augmentation framework for 3d object detection
topic 3D object detection
data augmentation
diversity
generalization
point cloud
robustness
url https://doi.org/10.1049/cit2.70001
work_keys_str_mv AT huaijinliu synauganeffectiveandgeneralsynchronousdataaugmentationframeworkfor3dobjectdetection
AT jixiangdu synauganeffectiveandgeneralsynchronousdataaugmentationframeworkfor3dobjectdetection
AT yongzhang synauganeffectiveandgeneralsynchronousdataaugmentationframeworkfor3dobjectdetection
AT hongbozhang synauganeffectiveandgeneralsynchronousdataaugmentationframeworkfor3dobjectdetection
AT jiandianzeng synauganeffectiveandgeneralsynchronousdataaugmentationframeworkfor3dobjectdetection