TY - JOUR
T1 - Key Point Sensitive Loss for Long-Tailed Visual Recognition
AU - Li, Mengke
AU - Cheung, Yiu-Ming
AU - Hu, Zhikai
N1 - Publisher copyright:
© 1979-2012 IEEE.
Funding information:
This work was supported in part by NSFC/Research Grants Council (RGC) Joint Research Scheme under Grant N_HKBU214/21, in part by General Research Fund of RGC under Grant 12201321, and in part by Hong Kong Baptist University under Grant RC-FNRA-IG/18-19/SCI/03.
PY - 2023/4/1
Y1 - 2023/4/1
N2 - For long-tailed distributed data, existing classification models often learn overwhelmingly on the head classes while ignoring the tail classes, resulting in poor generalization capability. To address this problem, we thereby propose a new approach in this paper, in which a key point sensitive (KPS) loss is presented to regularize the key points strongly to improve the generalization performance of the classification model. Meanwhile, in order to improve the performance on tail classes, the proposed KPS loss also assigns relatively large margins on tail classes. Furthermore, we propose a gradient adjustment (GA) optimization strategy to re-balance the gradients of positive and negative samples for each class. By virtue of the gradient analysis of the loss function, it is found that the tail classes always receive negative signals during training, which misleads the tail prediction to be biased towards the head. The proposed GA strategy can circumvent excessive negative signals on tail classes and further improve the overall classification accuracy. Extensive experiments conducted on long-tailed benchmarks show that the proposed method is capable of significantly improving the classification accuracy of the model in tail classes while maintaining competent performance in head classes.
AB - For long-tailed distributed data, existing classification models often learn overwhelmingly on the head classes while ignoring the tail classes, resulting in poor generalization capability. To address this problem, we thereby propose a new approach in this paper, in which a key point sensitive (KPS) loss is presented to regularize the key points strongly to improve the generalization performance of the classification model. Meanwhile, in order to improve the performance on tail classes, the proposed KPS loss also assigns relatively large margins on tail classes. Furthermore, we propose a gradient adjustment (GA) optimization strategy to re-balance the gradients of positive and negative samples for each class. By virtue of the gradient analysis of the loss function, it is found that the tail classes always receive negative signals during training, which misleads the tail prediction to be biased towards the head. The proposed GA strategy can circumvent excessive negative signals on tail classes and further improve the overall classification accuracy. Extensive experiments conducted on long-tailed benchmarks show that the proposed method is capable of significantly improving the classification accuracy of the model in tail classes while maintaining competent performance in head classes.
KW - Long-tailed classification
KW - class imbalance
KW - imbalance learning
KW - long-tailed visual recognition
UR - http://www.scopus.com/inward/record.url?scp=85135754303&partnerID=8YFLogxK
U2 - 10.1109/TPAMI.2022.3196044
DO - 10.1109/TPAMI.2022.3196044
M3 - Journal article
SN - 0162-8828
VL - 45
SP - 4812
EP - 4825
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
IS - 4
ER -