.png){kind=icon}
返回
Robust learning from noisy web data for fine-Grained recognition
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2022.109063.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Due to DNNs' memorization effect, label noise lessens the performance of the web-supervised fine-grained visual categorization task. Previous literature primarily relies on small-loss instances for sub-sequent training. The current state-of-the-art approach JoCoR additionally employs explicit consistency constraints to make clean samples more confident. However, a joint loss designed for both sample selec-tion criteria and parameter updating is not competent for training a robust model in the presence of web noise. Especially, false positives are assigned with larger weights, causing the model to pay more attention to misclassified noisy images. Besides, leveraging weight decay to forget discarded noisy instances is too slow and implicit to take effect. Therefore, we propose a simple yet effective approach named MS-DeJOR (Multi-Scale training with Decoupled Joint Optimization and Refurbishment). In contrast to JoCoR, we decouple sample selection from training procedure to handle the above problems. Specifically, a negative entropy term is applied to prevent false positives from being overemphasized. The model can explicitly forget those samples identified as noise by imposing such a regularization term on all training data. Fur-thermore, we use accumulated predictions to refurbish the noisy labels and re-weight training images to boost the model performance. A multi-scale feature enhancement module is adopted to extract discrimi-native and subtle feature representations. Extensive experiments show that MS-DeJOR yields state-of-the-art performances on three web-supervised fine-grained datasets, demonstrating the effectiveness of our approach. The data and source code have been available at https://github.com/msdejor/MS-DeJOR .(c) 2022 Elsevier Ltd. All rights reserved.
Keyword:
Fine-grained
Web-supervised
Noisy web data
Robust learning
期刊
IF:
7.6
论文数:
1.3W
被引数:
4.5W
机构
引用论文
LG-CNN: From local parts to global discrimination for fine-grained recognitionLg-cnn: 从局部到细粒度识别的全局歧视.png)
PATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
分享收藏
Mask-CNN: Localizing parts and selecting descriptors for fine-grained bird species categorizationPATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
分享收藏Co-LDL: A Co-Training-Based Label Distribution Learning Method for Tackling Label NoiseCo-ldl: 一种基于协同训练的标签分布学习方法，用于解决标签噪声
PREAI
分享收藏 没有更多内容
没有更多内容