Normalizing Batch Normalization for Long-Tailed Recognition
- 27 December 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 34 (10577149) , 209-220
Abstract
In real-world scenarios, the number of training samples across classes usually subjects to a long-tailed distribution. The conventionally trained network may achieve unexpected inferior performance on the rare class compared to the frequent class. Most previous works attempt to rectify the network bias from the data-level or from the classifier-level. Differently, in this paper, we identify that the bias towards the frequent class may be encoded into features, i.e., the rare-specific features which play a key role in discriminating the rare class are much weaker than the frequent-specific features. Based on such an observation, we introduce a simple yet effective approach, normalizing the parameters of Batch Normalization (BN) layer to explicitly rectify the feature bias. To achieve this end, we represent theWeight/Bias parameters of a BN layer as a vector, normalize it into a unit one and multiply the unit vector by a scalar learnable parameter. Through decoupling the direction and magnitude of parameters in BN layer to learn, the Weight/Bias exhibits a more balanced distribution and thus the strength of features becomes more even. Extensive experiments on various long-tailed recognition benchmarks (i.e., CIFAR-10/100-LT, ImageNet-LT and iNaturalist 2018) show that our method outperforms previous state-of-the-arts remarkably.Keywords
All Related Versions
Funding Information
- National Key Research and Development Program of China (2023YFC3306401, 2022YFB3104900)
- National Natural Science Foundation of China (62406298)
- Natural Science Foundation of Zhejiang Province (LD24F020007)
- Beijing Natural Science Foundation (L223024, L244043)
- “One Thousand Plan” projects in Jiangxi Province (Jxsq2023102268)
This publication has 37 references indexed in Scilit:
- MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognition2021
- Distribution Alignment: A Unified Framework for Long-tail Visual Recognition2021
- Rethinking Class-Balanced Methods for Long-Tailed Visual Recognition From a Domain Adaptation Perspective2020
- Equalization Loss for Long-Tailed Object Recognition2020
- Large-Scale Long-Tailed Recognition in an Open World2019
- Class-Balanced Loss Based on Effective Number of Samples2019
- The iNaturalist Species Classification and Detection Dataset2018
- Deep Over-sampling Framework for Classifying Imbalanced Data2017
- Aggregated Residual Transformations for Deep Neural Networks2017
- Deep Residual Learning for Image Recognition2016