Enhancing convolutional neural networks for face recognition with occlusion maps and batch triplet loss

Abstract

Despite the recent success of convolutional neural networks for computer vision applications, unconstrained face recognition remains a challenge. In this work, we make two contributions to the field. Firstly, we consider the problem of face recognition with partial occlusions and show how current approaches might suffer significant performance degradation when dealing with this kind of face images. We propose a simple method to find out which parts of the human face are more important to achieve a high recognition rate, and use that information during training to force a convolutional neural network to learn discriminative features from all the face regions more equally, including those that typical approaches tend to pay less attention to. We test the accuracy of the proposed method when dealing with real-life occlusions using the AR face database. Secondly, we propose a novel loss function called batch triplet loss that improves the performance of the triplet loss by adding an extra term to the loss function to cause minimisation of the standard deviation of both positive and negative scores. We show consistent improvement in the Labeled Faces in the Wild (LFW) benchmark by applying both proposed adjustments to the convolutional neural network training.

Prediction of monthly Arctic sea ice concentrations using satellite and reanalysis data based on convolutional neural networks

“…This is a useful characteristic when conducting a time-series analysis, and, thus, two SIC time-series climatology predictors (SIC 1 year before and SIC 1 month before) were used in this study. Although there is no clear physical explanation of why the interannual variations would contribute to the forecasting skill, it clearly worked well in long-term SIC forecasting in previous studies (Wang et al, 2016a;Chi and Kim, 2017). Furthermore, we used two supplementary predictors that indicate the anomalies of SIC 1 year before and SIC 1 month before in order to consider anomalous sea ice conditions in the models.…”

“…To prevent heavy computation, both the stride (i.e., how to shift a moving kernel) and the pooling (i.e., how to conduct downsampling) techniques are widely used, which make the size of the input data in the following convolutional process reduced. To avoid too much data reduction, many studies have adopted a padding technique, which covers input data with extra dummy values (Wang et al, 2016a). The feature map achieved through the convolutional process is a convolved map that contains a higher level of features of an image (Chen et al, 2015).…”

Prediction of monthly Arctic sea ice concentration using satellite and reanalysis data based on convolutional neural networks

“…Recent deep learning studies have tried to report explainable results using various visualization approaches such as heat maps 395 and occlusion maps (Brahimi et al, 2017;Trigueros et al, 2018). The present study explained the model using a variable sensitivity analysis, as well as the inspection of the spatial distribution.…”

Occlusion Robust Face Recognition Based on Mask Learning With Pairwise Differential Siamese Network

“…One possible approach to improve the performance of CNN models under partial occlusions is to train the network with occluded faces. Daniel et al [29] proposed to augment training data with synthetic occluded faces in a strategic manner and observed improved performance. However, it does not solve the problem intrinsically because it only ensures the features are more locally and equally extracted, as analyzed in [22].…”