GRAMO: geometric resampling augmentation for monocular 3D object detection

doi:10.1007/s11704-023-3242-2

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多模态人工智能系统全国重点实验室

	GRAMO: geometric resampling augmentation for monocular 3D object detection
	Guan, He 1,2; Song, Chunfeng 1,2; Zhang, Zhaoxiang 1,2
刊名	FRONTIERS OF COMPUTER SCIENCE
	2024-10-01
卷号	18 期号:5 页码:9
关键词	3D detection monocular augmentation geometry
ISSN号	2095-2228
DOI	10.1007/s11704-023-3242-2
通讯作者	Zhang, Zhaoxiang(zhaoxiang.zhang@ia.ac.cn)
英文摘要	Data augmentation is widely recognized as an effective means of bolstering model robustness. However, when applied to monocular 3D object detection, non-geometric image augmentation neglects the critical link between the image and physical space, resulting in the semantic collapse of the extended scene. To address this issue, we propose two geometric-level data augmentation operators named Geometric-Copy-Paste (Geo-CP) and Geometric-Crop-Shrink (Geo-CS). Both operators introduce geometric consistency based on the principle of perspective projection, complementing the options available for data augmentation in monocular 3D. Specifically, Geo-CP replicates local patches by reordering object depths to mitigate perspective occlusion conflicts, and Geo-CS re-crops local patches for simultaneous scaling of distance and scale to unify appearance and annotation. These operations ameliorate the problem of class imbalance in the monocular paradigm by increasing the quantity and distribution of geometrically consistent samples. Experiments demonstrate that our geometric-level augmentation operators effectively improve robustness and performance in the KITTI and Waymo monocular 3D detection benchmarks.
资助项目	National Key R&D Program of China[2022ZD0160102] ; National Natural Science Foundation of China[61836014] ; National Natural Science Foundation of China[U21B2042] ; National Natural Science Foundation of China[62072457] ; National Natural Science Foundation of China[62006231]
WOS研究方向	Computer Science
语种	英语
出版者	HIGHER EDUCATION PRESS
WOS记录号	WOS:001142745300001
资助机构	National Key R&D Program of China ; National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/54816]
专题	多模态人工智能系统全国重点实验室
通讯作者	Zhang, Zhaoxiang
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Ctr Res Intelligent Percept & Comp, State Key Lab Multimodal Artificial Intelligence S, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Guan, He,Song, Chunfeng,Zhang, Zhaoxiang. GRAMO: geometric resampling augmentation for monocular 3D object detection[J]. FRONTIERS OF COMPUTER SCIENCE,2024,18(5):9.
APA	Guan, He,Song, Chunfeng,&Zhang, Zhaoxiang.(2024).GRAMO: geometric resampling augmentation for monocular 3D object detection.FRONTIERS OF COMPUTER SCIENCE,18(5),9.
MLA	Guan, He,et al."GRAMO: geometric resampling augmentation for monocular 3D object detection".FRONTIERS OF COMPUTER SCIENCE 18.5(2024):9.