一种基于角度距离损失函数和卷积神经网络的人脸识别算法 下载: 1819次
A Face Recognition Algorithm Based on Angular Distance Loss Function and Convolutional Neural Network
天津大学电气自动化与信息工程学院, 天津 300072
图 & 表
图 1. 人脸识别测试协议的比较。(a)闭集人脸识别;(b)开集人脸识别
Fig. 1. Comparison of test protocol of face recognition. (a) Closed-set face recognition; (b) open-set face recognition
下载图片 查看原文
图 2. softmax损失函数的比较。(a)传统的softmax损失函数;(b)改进的softmax损失函数
Fig. 2. Comparison of softmax loss function. (a) Traditional softmax loss function; (b) improved softmax loss function
下载图片 查看原文
图 3. 本文提出的基于角度距离损失函数的示意图
Fig. 3. Schematic of the proposed angular distance loss function
下载图片 查看原文
图 4. 密集连接网络的结构
Fig. 4. Structure of densely connected networks
下载图片 查看原文
图 5. 激活函数的对比。(a) ReLU;(b) PReLU
Fig. 5. Comparison of activation functions. (a) ReLU; (b) PReLU
下载图片 查看原文
图 6. 网络整体结构
Fig. 6. Integral structure of network
下载图片 查看原文
图 7. 不同超参数ω的人脸识别准确率
Fig. 7. Face recognition accuracy versus hyperparameter ω
下载图片 查看原文
图 8. 不同层数和不同损失函数的网络结构在LFW数据集上的测试准确率
Fig. 8. Test accuracy of LFW dataset for network structures with different layer numbers and different loss functions
下载图片 查看原文
图 9. 不同层数和宽度的网络结构在LFW数据集上的测试准确率
Fig. 9. Test accuracy of LFW dataset for network structures with different layer numbers and widths
下载图片 查看原文
图 10. 本文实验实施流程
Fig. 10. Proposed implementation process
下载图片 查看原文
表 1三种损失函数的分类边界对比
Table1. Comparison of classification boundaries of loss functions
Loss function | Decision boundary |
---|
Original softmax loss | (W1-W2)x+b1-b2=0 | Modified softmax loss | (cosθ1-cosθ2)=0 | Angular distance loss | {cosθ1-cos[(1-ω)θ2]}=0 for class 1 | {cos[(1-ω)θ1]-cosθ2}=0 for class 2 |
|
查看原文
表 2密集连接结构的具体配置
Table2. Specific configuration of the dense connection structure
Layer | Output size | DenseFace-42 | DenseFace-54 | DenseFace-78 | DenseFace-122 |
---|
Dense block 1 | 56×56 | ×4 | ×6 | ×6 | ×6 | Dense block 2 | 28×28 | ×5 | ×6 | ×12 | ×12 | Dense block 3 | 14×14 | ×5 | ×6 | ×12 | ×24 | Dense block 4 | 7×7 | ×4 | ×6 | ×6 | ×16 |
|
查看原文
表 3几种CNN模型参数量的比较
Table3. Comparison of parameter quantities of several convolutional neural network models
Net structure | Input size /pixel | Depth /layer | Parameter /106 |
---|
LeNet | 32×32×1 | 5 | 0.062 | AlexNet | 227×227×3 | 8 | 62.4 | VGGNet | 224×224×3 | 16 | 138.4 | GoogleNet | 224×224×3 | 22 | 5.3 | ResNet | 224×224×3 | 152 | 61.3 | DenseFace (width: 32) | 112×112×3 | 42 | 6.7 | 54 | 7.3 | 78 | 8.9 | 122 | 12.8 | DenseFace (width: 16) | 112×112×3 | 42 | 5.78 | 54 | 5.9 | 78 | 6.37 | 122 | 7.4 |
|
查看原文
表 4不同损失函数、人脸识别算法的测试准确率
Table4. Test accuracy of different loss functions or face recognition algorithms
Method | Dataset | Data amount /106 | Accuracy /% |
---|
DeepFace | LFW | 4 | 97.33 | FaceNet | LFW | 200 | 99.67 | Deep FR | LFW | 2.6 | 98.85 | DeepID2+ | LFW | 0.3 | 98.74 | Center Face | LFW | 0.7 | 99.31 | Softmax loss | CAISA-WebFace | 0.49 | 97.78 | Triplet loss | CAISA-WebFace | 0.49 | 98.65 | Center loss | CAISA-WebFace | 0.49 | 99.02 | L-softmax loss | CAISA-WebFace | 0.49 | 99.15 | Angular distance loss | CAISA-WebFace | 0.49 | 99.45 |
|
查看原文
表 5不同损失函数、人脸识别算法在MegaFace数据集上的测试准确率
Table5. Test accuracy of different loss functions or face recognition algorithms on the MegaFace dataset
Method | Test protocol | Accuracy /% |
---|
Face identification | Face verification |
---|
FaceNet | large | 70.496 | 86.473 | Deepsense | large | 74.798 | 87.764 | Deepsense | small | 70.983 | 82.851 | Softmax loss | small | 54.628 | 65.732 | Triplet loss | small | 64.698 | 78.030 | Center loss | small | 65.334 | 80.106 | L-softmax loss | small | 67.035 | 80.185 | Angular softmax loss | small | 72.534 | 85.348 |
|
查看原文
龙鑫, 苏寒松, 刘高华, 陈震宇. 一种基于角度距离损失函数和卷积神经网络的人脸识别算法[J]. 激光与光电子学进展, 2018, 55(12): 121505. Xin Long, Hansong Su, Gaohua Liu, Zhenyu Chen. A Face Recognition Algorithm Based on Angular Distance Loss Function and Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2018, 55(12): 121505.