TY - GEN
T1 - Class Similarity Weighted Knowledge Distillation for Continual Semantic Segmentation
AU - Phan, Minh Hieu
AU - Ta, The Anh
AU - Phung, Son Lam
AU - Tran-Thanh, Long
AU - Bouzerdoum, Abdesselam
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Deep learning models are known to suffer from the problem of catastrophic forgetting when they incrementally learn new classes. Continual learning for semantic segmentation (CSS) is an emerging field in computer vision. We identify a problem in CSS: A model tends to be confused between old and new classes that are visually similar, which makes it forget the old ones. To address this gap, we propose REMINDER - a new CSS framework and a novel class similarity knowledge distillation (CSW-KD) method. Our CSW-KD method distills the knowledge of a previous model on old classes that are similar to the new one. This provides two main benefits: (i) selectively revising old classes that are more likely to be forgotten, and (ii) better learning new classes by relating them with the previously seen classes. Extensive experiments on Pascal-Voc 2012 and ADE20k datasets show that our approach outperforms state-of-the-art methods on standard CSS settings by up to 7.07% and 8.49%, respectively.
AB - Deep learning models are known to suffer from the problem of catastrophic forgetting when they incrementally learn new classes. Continual learning for semantic segmentation (CSS) is an emerging field in computer vision. We identify a problem in CSS: A model tends to be confused between old and new classes that are visually similar, which makes it forget the old ones. To address this gap, we propose REMINDER - a new CSS framework and a novel class similarity knowledge distillation (CSW-KD) method. Our CSW-KD method distills the knowledge of a previous model on old classes that are similar to the new one. This provides two main benefits: (i) selectively revising old classes that are more likely to be forgotten, and (ii) better learning new classes by relating them with the previously seen classes. Extensive experiments on Pascal-Voc 2012 and ADE20k datasets show that our approach outperforms state-of-the-art methods on standard CSS settings by up to 7.07% and 8.49%, respectively.
KW - Computer vision theory
KW - Deep learning architectures and techniques
KW - Efficient learning and inferences
KW - Representation learning
KW - Scene analysis and understanding
KW - Segmentation
KW - Vision applications and systems
KW - grouping and shape analysis
UR - http://www.scopus.com/inward/record.url?scp=85141768611&partnerID=8YFLogxK
U2 - 10.1109/CVPR52688.2022.01636
DO - 10.1109/CVPR52688.2022.01636
M3 - Conference contribution
AN - SCOPUS:85141768611
T3 - Ieee Conference On Computer Vision And Pattern Recognition
SP - 16845
EP - 16854
BT - 2022 Ieee/cvf Conference On Computer Vision And Pattern Recognition (cvpr 2022)
PB - IEEE Computer Society
T2 - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
Y2 - 19 June 2022 through 24 June 2022
ER -