SIA OpenIR  > 机器人学研究室
I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting
Dong JH(董家华)1,2,3; Cong Y(丛杨)1,2; Sun G(孙干)1,2; Ma BT(马兵涛)1,2,3; Wang LC(王莅尘)4
Department机器人学研究室
Conference Name35th AAAI Conference on Artificial Intelligence / 33rd Conference on Innovative Applications of Artificial Intelligence / 11th Symposium on Educational Advances in Artificial Intelligence
Conference DateFebuary 2-9, 2021
Conference PlaceELECTR NETWORK
Author of SourceAssociation for the Advancement of Artificial Intelligence
Source PublicationTHIRTY-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
PublisherAAAI
Publication PlacePalo Alto, California
2021
Pages6066-6074
Indexed ByCPCI(ISTP)
WOS IDWOS:000680423506020
Contribution Rank1
ISSN2159-5399
ISBN978-1-57735-866-4
Abstract3D 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.
Language英语
Citation statistics
Document Type会议论文
Identifierhttp://ir.sia.cn/handle/173321/29553
Collection机器人学研究室
Affiliation1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
2.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, 110016, China
3.University of Chinese Academy of Sciences, Beijing, 100049, China
4.Northeastern University, Boston, USA
Recommended Citation
GB/T 7714
Dong JH,Cong Y,Sun G,et al. I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting[C]//Association for the Advancement of Artificial Intelligence. Palo Alto, California:AAAI,2021:6066-6074.
Files in This Item:
File Name/Size DocType Version Access License
I3DOL_ Incremental 3(1954KB)会议论文 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Dong JH(董家华)]'s Articles
[Cong Y(丛杨)]'s Articles
[Sun G(孙干)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Dong JH(董家华)]'s Articles
[Cong Y(丛杨)]'s Articles
[Sun G(孙干)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Dong JH(董家华)]'s Articles
[Cong Y(丛杨)]'s Articles
[Sun G(孙干)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: I3DOL_ Incremental 3D Object Learning without Catastrophic Forgetting.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.