I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting

CORC > 沈阳自动化研究所 > 中国科学院沈阳自动化研究所 > 机器人学研究室

	I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting
	Dong JH(董家华)1,2,4; Cong Y(丛杨)1,4; Sun G(孙干)1,4; Ma BT(马兵涛)1,2,4; Wang LC(王莅尘)3
	2021
会议日期	Febuary 2-9, 2021
会议地点	ELECTR NETWORK
页码	6066-6074
英文摘要	3D object classification has attracted appealing attentions in academic researches and industrial applications. However, most existing methods need to access the training data of past 3D object classes when facing the common real-world scenario: new classes of 3D objects arrive in a sequence. Moreover, the performance of advanced approaches degrades dramatically for past learned classes (i.e., catastrophic forgetting), due to the irregular and redundant geometric structures of 3D point cloud data. To address these challenges, we propose a new Incremental 3D Object Learning (i.e., I3DOL) model, which is the first exploration to learn new classes of 3D object continually. Specifically, an adaptive-geometric centroid module is designed to construct discriminative local geometric structures, which can better characterize the irregular point cloud representation for 3D object. Afterwards, to prevent the catastrophic forgetting brought by redundant geometric information, a geometric-aware attention mechanism is developed to quantify the contributions of local geometric structures, and explore unique 3D geometric characteristics with high contributions for classes incremental learning. Meanwhile, a score fairness compensation strategy is proposed to further alleviate the catastrophic forgetting caused by unbalanced data between past and new classes of 3D object, by compensating biased prediction for new classes in the validation phase. Experiments on 3D representative datasets validate the superiority of our I3DOL framework.
源文献作者	Association for the Advancement of Artificial Intelligence
产权排序	1
会议录	THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE
会议录出版者	AAAI
会议录出版地	Palo Alto, California
语种	英语
ISSN号	2159-5399
ISBN号	978-1-57735-866-4
WOS记录号	WOS:000680423506020
内容类型	会议论文
源URL	[http://ir.sia.cn/handle/173321/29553]
专题	沈阳自动化研究所_机器人学研究室
通讯作者	Cong Y(丛杨)
作者单位	1.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, 110016, China 2.University of Chinese Academy of Sciences, Beijing, 100049, China 3.Northeastern University, Boston, USA 4.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
推荐引用方式 GB/T 7714	Dong JH,Cong Y,Sun G,et al. I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting[C]. 见:. ELECTR NETWORK. Febuary 2-9, 2021.