L3DOC: Lifelong 3D Object Classification | |
Liu YY(刘宇阳)1,2,3; Cong Y(丛杨)1,2; Sun G(孙干)1,2; Zhang T(张涛)1,2,3; Dong JH(董家华)1,2,3; Liu HS(刘洪森)4 | |
刊名 | IEEE TRANSACTIONS ON IMAGE PROCESSING |
2021 | |
卷号 | 30页码:7486-7498 |
关键词 | 3D object classification lifelong learning point-knowledge task-relevant knowledge distillation |
ISSN号 | 1057-7149 |
产权排序 | 1 |
英文摘要 | 3D object classification has been widely applied in both academic and industrial scenarios. However, most state-of-the-art algorithms rely on a fixed object classification task set, which cannot tackle the scenario when a new 3D object classification task is coming. Meanwhile, the existing lifelong learning models can easily destroy the learned tasks performance, due to the unordered, large-scale, and irregular 3D geometry data. To address these challenges, we propose a Lifelong 3D Object Classification (i.e., L3DOC) model, which can consecutively learn new 3D object classification tasks via imitating human learning. More specifically, the core idea of our model is to capture and store the cross-task common knowledge of 3D geometry data in a 3D neural network, named as point-knowledge, through employing layer-wise point-knowledge factorization architecture. Afterwards, a task-relevant knowledge distillation mechanism is employed to connect the current task to previous relevant tasks and effectively prevent catastrophic forgetting. It consists of a point-knowledge distillation module and a transforming-space distillation module, which transfers the accumulated point-knowledge from previous tasks and soft-transfers the compact factorized representations of the transforming-space, respectively. To our best knowledge, the proposed L3DOC algorithm is the first attempt to perform deep learning on 3D object classification tasks in a lifelong learning way. Extensive experiments on several point cloud benchmarks illustrate the superiority of our L3DOC model over the state-of-the-art lifelong learning methods. |
资助项目 | New Generation of Artificial Intelligence[2018AAA0102905] ; National Natural Science Foundation[61821005] ; National Natural Science Foundation[62003336] ; State Key Laboratory of Robotics[2022-Z06] ; National Postdoctoral Innovative Talents Support Program[BX20200353] |
WOS关键词 | EFFICIENT |
WOS研究方向 | Computer Science ; Engineering |
语种 | 英语 |
WOS记录号 | WOS:000692208400006 |
资助机构 | New Generation of Artificial Intelligence [2018AAA0102905] ; National Natural Science FoundationNational Natural Science Foundation of China (NSFC) [61821005, 62003336] ; State Key Laboratory of Robotics [2022-Z06] ; National Postdoctoral Innovative Talents Support Program [BX20200353] |
内容类型 | 期刊论文 |
源URL | [http://ir.sia.cn/handle/173321/29576] |
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.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, China 3.University of Chinese Academy of Sciences, Beijing, China 4.JD.com, Inc., Beijing, China |
推荐引用方式 GB/T 7714 | Liu YY,Cong Y,Sun G,et al. L3DOC: Lifelong 3D Object Classification[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2021,30:7486-7498. |
APA | Liu YY,Cong Y,Sun G,Zhang T,Dong JH,&Liu HS.(2021).L3DOC: Lifelong 3D Object Classification.IEEE TRANSACTIONS ON IMAGE PROCESSING,30,7486-7498. |
MLA | Liu YY,et al."L3DOC: Lifelong 3D Object Classification".IEEE TRANSACTIONS ON IMAGE PROCESSING 30(2021):7486-7498. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论