TY - GEN
T1 - Incremental object learning from contiguous views
AU - Stojanov, Stefan
AU - Mishra, Samarth
AU - Thai, Ngoc Anh
AU - Dhanda, Nikhil
AU - Humayun, Ahmad
AU - Yu, Chen
AU - Smith, Linda B.
AU - Rehg, James M.
N1 - Funding Information:
We would like to thank all the reviewers and Qian Shao for his early contributions to the incremental learning experiments. This work was supported by NSF Awards BCS-1524565 and BCS-1523982. In addition, this work was partially supported by the Indiana University Areas of Emergent Research initiative in Learning: Brains, Machines, Children.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/6
Y1 - 2019/6
N2 - In this work, we present CRIB (Continual Recognition Inspired by Babies), a synthetic incremental object learning environment that can produce data that models visual imagery produced by object exploration in early infancy. CRIB is coupled with a new 3D object dataset, Toys-200, that contains 200 unique toy-like object instances, and is also compatible with existing 3D datasets. Through extensive empirical evaluation of state-of-the-art incremental learning algorithms, we find the novel empirical result that repetition can significantly ameliorate the effects of catastrophic forgetting. Furthermore, we find that in certain cases repetition allows for performance approaching that of batch learning algorithms. Finally, we propose an unsupervised incremental learning task with intriguing baseline results.
AB - In this work, we present CRIB (Continual Recognition Inspired by Babies), a synthetic incremental object learning environment that can produce data that models visual imagery produced by object exploration in early infancy. CRIB is coupled with a new 3D object dataset, Toys-200, that contains 200 unique toy-like object instances, and is also compatible with existing 3D datasets. Through extensive empirical evaluation of state-of-the-art incremental learning algorithms, we find the novel empirical result that repetition can significantly ameliorate the effects of catastrophic forgetting. Furthermore, we find that in certain cases repetition allows for performance approaching that of batch learning algorithms. Finally, we propose an unsupervised incremental learning task with intriguing baseline results.
KW - Categorization
KW - Datasets and Evaluation
KW - Deep Learning
KW - Image and Video Synthesis
KW - Recognition: Detection
KW - Retrieval
UR - http://www.scopus.com/inward/record.url?scp=85078564653&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85078564653&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2019.00898
DO - 10.1109/CVPR.2019.00898
M3 - Conference contribution
AN - SCOPUS:85078564653
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 8769
EP - 8778
BT - Proceedings - 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2019
PB - IEEE Computer Society
T2 - 32nd IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2019
Y2 - 16 June 2019 through 20 June 2019
ER -