IROS 2019 Lifelong Robotic Vision: Object Recognition Challenge [Competitions]

Bae, Heechul; Brophy, Eoin; Chan, Rosa H. M.; Chen, Baoquan; Feng, Fan; Graffieti, Gabriele; Goel, Vidit; Hao, Xinyue; Han, Hongbin; Kanagarajah, Sathursan; Kumar, Somesh; Lam, Siew-Kei; Lam, Tsun Ip; Lan, Chuanlin; Liu, Qi; Lomonaco, Vincenzo; Ma, Liang; Maltoni, Davide; Parisi, German I.; Pellegrini, Lorenzo; Piyasena, Duvindu; Pu, Shiliang; She, Qi; Sheet, Debdoot; Song, Soonyong; Son, Young-Sung; Wang, Zhengwei; Ward, Tomás E.; Wu, Jianwen; Wu, Meiqing; Xie, Dong; Xu, Yangsheng; Yang, Lin; Yang, Qihan; Zhong, Qiaoyong; Zhou, Liguang

doi:10.1109/mra.2020.2987186

Cited by 10 publications

(2 citation statements)

References 17 publications

(18 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the research of incremental learning is very important to robotics, there are also some datasets proposed for robotics. In IROS 2019-Lifelong Robotic Vision Competition, OpenLORIS-Object was proposed to promote lifelong learning research and applications in the field of robot vision, including daily necessities in homes, offices, campuses and shopping malls [ 121 , 131 ]. The dataset clearly quantifies the illumination, occlusion, object size, camera-object distance/angle, and clutter.…”

Section: Datasetsmentioning

confidence: 99%

An Appraisal of Incremental Learning Methods

Luo

Yin

Bai

et al. 2020

Entropy

View full text Add to dashboard Cite

As a special case of machine learning, incremental learning can acquire useful knowledge from incoming data continuously while it does not need to access the original data. It is expected to have the ability of memorization and it is regarded as one of the ultimate goals of artificial intelligence technology. However, incremental learning remains a long term challenge. Modern deep neural network models achieve outstanding performance on stationary data distributions with batch training. This restriction leads to catastrophic forgetting for incremental learning scenarios since the distribution of incoming data is unknown and has a highly different probability from the old data. Therefore, a model must be both plastic to acquire new knowledge and stable to consolidate existing knowledge. This review aims to draw a systematic review of the state of the art of incremental learning methods. Published reports are selected from Web of Science, IEEEXplore, and DBLP databases up to May 2020. Each paper is reviewed according to the types: architectural strategy, regularization strategy and rehearsal and pseudo-rehearsal strategy. We compare and discuss different methods. Moreover, the development trend and research focus are given. It is concluded that incremental learning is still a hot research area and will be for a long period. More attention should be paid to the exploration of both biological systems and computational models.

show abstract

Section: Datasetsmentioning

confidence: 99%

An Appraisal of Incremental Learning Methods

Luo

Yin

Bai

et al. 2020

Entropy

View full text Add to dashboard Cite

show abstract

“…To solve these problems, we redefine the CNN-based superpixel segmentation as a lifelong learning task [22,9,4] which can sequentially learn a unified model online. In addition, a lightweight unsupervised CNN-based superpixel segmentation method called LNS-Net is proposed to learn superpixel in a non-iterative and lifelong manner.…”

Section: Fem Ncm-sel Ncm-clmentioning

confidence: 99%

Learning the Superpixel in a Non-iterative and Lifelong Manner

Zhu¹,

She²,

Zhang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Superpixel is generated by automatically clustering pixels in an image into hundreds of compact partitions, which is widely used to perceive the object contours for its excellent contour adherence. Although some works use the Convolution Neural Network (CNN) to generate high-quality superpixel, we challenge the design principles of these networks, specifically for their dependence on manual labels and excess computation resources, which limits their flexibility compared with the traditional unsupervised segmentation methods. We target at redefining the CNN-based superpixel segmentation as a lifelong clustering task and propose an unsupervised CNN-based method called LNS-Net. The LNS-Net can learn superpixel in a non-iterative and lifelong manner without any manual labels. Specifically, a lightweight feature embedder is proposed for LNS-Net to efficiently generate the cluster-friendly features. With those features, seed nodes can be automatically assigned to cluster pixels in a non-iterative way. Additionally, our LNS-Net can adapt the sequentially lifelong learning by rescaling the gradient of weight based on both channel and spatial context to avoid overfitting. Experiments show that the proposed LNS-Net achieves significantly better performance on three benchmarks with nearly ten times lower complexity compared with other state-of-the-art methods.

show abstract