Unequal-Training for Deep Face Recognition With Long-Tailed Noisy Data

Zhong, Yaoyao; Deng, Weihong; Wang, Mei; Hu, Jiani; Jiang, Peng; Tao, Xunqiang; Huang, Yaohai

doi:10.1109/cvpr.2019.00800

Cited by 134 publications

(79 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They suggested that as the training proceeds, the model becomes more accurate and its predictions can be weighted more strongly, thereby gradually forgetting the original incorrect labels. Zhong et al (2019) used a similar approach for face identification. They first trained their model on a small dataset with less label noise and then fine-tuned it on data with stronger label noise using an iterative label update strategy similar to that explained above.…”

Section: Training Proceduresmentioning

confidence: 99%

Deep learning with noisy labels: Exploring techniques and remedies in medical image analysis

Karimi

Dou

Warfield

et al. 2020

Medical Image Analysis

477

234

View full text Add to dashboard Cite

Supervised training of deep learning models requires large labeled datasets. There is a growing interest in obtaining such datasets for medical image analysis applications. However, the impact of label noise has not received sufficient attention. Recent studies have shown that label noise can significantly impact the performance of deep learning models in many machine learning and computer vision applications. This is especially concerning for medical applications, where datasets are typically small, labeling requires domain expertise and suffers from high inter-and intra-observer variability, and erroneous predictions may influence decisions that directly impact human health. In this paper, we first review the state-of-the-art in handling label noise in deep learning. Then, we review studies that have dealt with label noise in deep learning for medical image analysis. Our review shows that recent progress on handling label noise in deep learning has gone largely unnoticed by the medical image analysis community. To help achieve a better understanding of the extent of the problem and its potential remedies, we conducted experiments with three medical imaging datasets with different types of label noise, where we investigated several existing strategies and developed new methods to combat the negative effect of label noise. Based on the results of these experiments and our review of the literature, we have made recommendations on methods that can be used to alleviate the effects of different types of label noise on deep models trained for medical image analysis. We hope that this article helps the medical image analysis researchers and developers in choosing and devising new techniques that effectively handle label noise in deep learning.

show abstract

Section: Training Proceduresmentioning

confidence: 99%

Deep learning with noisy labels: Exploring techniques and remedies in medical image analysis

Karimi

Dou

Warfield

et al. 2020

Medical Image Analysis

477

234

View full text Add to dashboard Cite

show abstract

“…In addition, the results of CE loss and CI loss are very close. When the ID is not 0, just like data distribution [1,2,3,4,5] and [2,4,6,8,10], the recognition rates are 0.535 & 0.587 and 0.564 & 0.601 in CE loss and CI loss, respectively.…”

Section: ) Results For Imbalanced Image Classification: Results On Tmentioning

confidence: 99%

“…A practical recognition system must handle class imbalanced data. However, the problem of learning from the class imbalanced dataset, i.e., the imbalanced learning problem, has been a challenging and longstanding problem [3]- [6]. Despite significant progress brought by deep learning [7]- [9], most of the existing deep learning methods consider class balanced datasets (such as the ImageNet1000 for image classification competition [10]) or moderately imbalanced datasets.…”

mentioning

confidence: 99%

A Class Imbalance Loss for Imbalanced Object Recognition

Zhang

Quan

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

The class imbalance problem exists widely in vision data. In these imbalanced datasets, the majority classes dominate the loss and influence the gradient. Hence, these datasets have a significantly negative impact on the performance of many stateof-the-art methods. In this article, we propose a class imbalance loss (CI loss) to handle this problem. To distinguish imbalanced datasets in accordance with the extent of imbalance, we also define an imbalance degree that works as a decision index factor in the CI loss. Because the minority classes with fewer samples probably lose chances in descending the gradient in the training process, CI loss is introduced to make these minority classes descend further than the majority classes. In view of the imbalanced distribution of data in few-shot learning, a method for generating an imbalanced few-shot learning dataset is presented in this article. We conducted a large number of experiments in the MiniImageNet dataset, which showed the effectiveness of an algorithm for model-agnostic metalearning for rapid adaptation with CI loss. In the problem of detecting 15 ship categories, our loss function is transplanted to a rotational region convolutional neural network detection method and a cascade network architecture and achieves higher mean average precision than focal loss and cross-entropy loss. In addition, the Mixed National Institute of Standards and Technology dataset and the Moving and Stationary Target Acquisition and Recognition dataset are sampled to imbalance datasets to verify the effectiveness of CI loss.

show abstract

“…Using a deep convolutional neural network (DCNN) for feature extraction for face representation is the preferred method of facial recognition [28,29]. DCNN performs a mapping operation on face images in a feature space with a small intra-class distance and a large inter-class distance.…”

Section: Softmax Loss Functionmentioning

confidence: 99%

MFCosface: A Masked-Face Recognition Algorithm Based on Large Margin Cosine Loss

et al. 2021

View full text Add to dashboard Cite

The world today is being hit by COVID-19. As opposed to fingerprints and ID cards, facial recognition technology can effectively prevent the spread of viruses in public places because it does not require contact with specific sensors. However, people also need to wear masks when entering public places, and masks will greatly affect the accuracy of facial recognition. Accurately performing facial recognition while people wear masks is a great challenge. In order to solve the problem of low facial recognition accuracy with mask wearers during the COVID-19 epidemic, we propose a masked-face recognition algorithm based on large margin cosine loss (MFCosface). Due to insufficient masked-face data for training, we designed a masked-face image generation algorithm based on the detection of the detection of key facial features. The face is detected and aligned through a multi-task cascaded convolutional network; and then we detect the key features of the face and select the mask template for coverage according to the positional information of the key features. Finally, we generate the corresponding masked-face image. Through analysis of the masked-face images, we found that triplet loss is not applicable to our datasets, because the results of online triplet selection contain fewer mask changes, making it difficult for the model to learn the relationship between mask occlusion and feature mapping. We use a large margin cosine loss as the loss function for training, which can map all the feature samples in a feature space with a smaller intra-class distance and a larger inter-class distance. In order to make the model pay more attention to the area that is not covered by the mask, we designed an Att-inception module that combines the Inception-Resnet module and the convolutional block attention module, which increases the weight of any unoccluded area in the feature map, thereby enlarging the unoccluded area’s contribution to the identification process. Experiments on several masked-face datasets have proved that our algorithm greatly improves the accuracy of masked-face recognition, and can accurately perform facial recognition with masked subjects.

show abstract

Unequal-Training for Deep Face Recognition With Long-Tailed Noisy Data

Cited by 134 publications

References 18 publications

Deep learning with noisy labels: Exploring techniques and remedies in medical image analysis

Deep learning with noisy labels: Exploring techniques and remedies in medical image analysis

A Class Imbalance Loss for Imbalanced Object Recognition

MFCosface: A Masked-Face Recognition Algorithm Based on Large Margin Cosine Loss

Contact Info

Product

Resources

About