Alexandridis, Konstantinos Panagiotis, Luo, Shan 
ORCID: 0000-0003-4760-0372, Nguyen, Anh ORCID: 0000-0002-1449-211X, Deng, Jiankang and Zafeiriou, Stefanos
(2023)
Inverse Image Frequency for Long-Tailed Image Recognition.
IEEE Transactions on Image Processing, 32.
pp. 5721-5736.
|
PDF
Inverse_Image_Frequency.pdf - Author Accepted Manuscript Download (10MB) | Preview |
Abstract
The long-tailed distribution is a common phenomenon in the real world. Extracted large scale image datasets inevitably demonstrate the long-tailed property and models trained with imbalanced data can obtain high performance for the over-represented categories, but struggle for the under-represented categories, leading to biased predictions and performance degradation. To address this challenge, we propose a novel de-biasing method named Inverse Image Frequency (IIF). IIF is a multiplicative margin adjustment transformation of the logits in the classification layer of a convolutional neural network. Our method achieves stronger performance than similar works and it is especially useful for downstream tasks such as long-tailed instance segmentation as it produces fewer false positive detections. Our extensive experiments show that IIF surpasses the state of the art on many long-tailed benchmarks such as ImageNet-LT, CIFAR-LT, Places-LT and LVIS, reaching 55.8% top-1 accuracy with ResNet50 on ImageNet-LT and 26.3% segmentation AP with MaskRCNN ResNet50 on LVIS. Code available at https://github.com/kostas1515/iif.
| Item Type: | Article |
|---|---|
| Divisions: | Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 26 Oct 2023 10:03 |
| Last Modified: | 21 Nov 2023 15:04 |
| DOI: | 10.1109/tip.2023.3321461 |
| Related URLs: | |
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3176456 |
Dimensions
Dimensions
