Improved Classroom Face Recognition Algorithm Based on InsightFace and Its Application

Xichu Tian; Hansong Su; Gaohua Liu; Tengteng Liu

doi:10.3788/LOP57.221501

Journals >Laser & Optoelectronics Progress >Volume 57 >Issue 22 >Page 221501 > Article

Laser & Optoelectronics Progress
Vol. 57, Issue 22, 221501 (2020)

Improved Classroom Face Recognition Algorithm Based on InsightFace and Its Application

Xichu Tian, Hansong Su, Gaohua Liu^*, and Tengteng Liu

Author Affiliations

School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China

show less

DOI: 10.3788/LOP57.221501 Cite this Article Set citation alerts

Xichu Tian, Hansong Su, Gaohua Liu, Tengteng Liu. Improved Classroom Face Recognition Algorithm Based on InsightFace and Its Application[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221501 Copy Citation Text

show less

Fig. 1. Mobile-Block structure with stride of 1.(a) MobileFaceNet; (b) Dual-MobileFaceNet

Download full size

Fig. 2. Mobile-Block structure with stride of 2. (a) MobileFaceNet; (b) Dual-MobileFaceNet

Download full size

Fig. 3. Schematic of Dual-MobileFaceNet structure

Download full size

Fig. 4. Schematic of double classifier structure

Download full size

Fig. 5. Examples of self-made training dataset

Download full size

Fig. 6. Classroom scene. (a) Real scene; (b) sketch map

Download full size

Fig. 7. Interface connection of Jetson TX2

Download full size

Fig. 8. Recognition results of proposed algorithm. (a) 8-people video; (b)16-people video

Download full size

Fig. 9. Recognition accuracy confusion matrix of 8-people video. (a) InsightFace; (b) Double classifier

Download full size

Fig. 10. Diagram of different face sizes. (a) Big face; (b) medium face; (c) small face

Download full size

Fig. 11. Recognition accuracy of different networks for different sizes of faces

Download full size

Fig. 12. Recognition accuracy of different algorithms for different sizes of faces

Download full size

Input size/Numberof channels	Type	Output size/Numberof channels	Operation	s	n	Pad
112×112/3	Convolution	56×56/64	3×3 Conv	2	1	1
56×56/64	Convolution	56×56/64	3×3 dw_Conv	1	1	1
56×56/64	Dual-Block	56×56/128+2k	$[\begin{array}{l} 1 \times 1 pw_Conv \\ 3 \times 3 dw_Conv \\ 1 \times 1 Linear_pw_Conv \end{array}]$	1	2
56×56/128+2k	Mobile-Block	28×28/64	$[\begin{array}{l} 1 \times 1 pw_Conv \\ 3 \times 3 dw_Conv \\ 1 \times 1 Linear_pw_Conv \end{array}]$	2	1	1
28×28/64	Dual-Block	28×28/128+6k	$[\begin{array}{l} 1 \times 1 pw_Conv \\ 3 \times 3 dw_Conv \\ 1 \times 1 Linear_pw_Conv \end{array}]$	1	6
28×28/128+6k	Mobile-Block	14×14/128	$[\begin{array}{l} 1 \times 1 pw_Conv \\ 3 \times 3 dw_Conv \\ 1 \times 1 Linear_pw_Conv \end{array}]$	2	1	1
14×14/128	Dual-Block	14×14/256+4k	$[\begin{array}{l} 1 \times 1 pw_Conv \\ 3 \times 3 dw_Conv \\ 1 \times 1 Linear_pw_Conv \end{array}]$	1	4
14×14/256+4k	Mobile-Block	7×7/128	$[\begin{array}{l} 1 \times 1 pw_Conv \\ 3 \times 3 dw_Conv \\ 1 \times 1 Linear_pw_Conv \end{array}]$	2	1	1
7×7/128	Dual-Block	7×7/256+2k	$[\begin{array}{l} 1 \times 1 pw_Conv \\ 3 \times 3 dw_Conv \\ 1 \times 1 Linear_pw_Conv \end{array}]$	1	2
7×7/256+2k	Convolution	7×7/512	1×1 pw_Conv	1	1	0
7×7/512	Convolution	1×1/512	7×7 Linear_Conv	1	1	0
1×1/512	Convolution	1×1/128	1×1 Linear_pw_Conv	1	1	0

Table 1. Network structure of Dual-MobileFaceNet

Network	Recognition accuracy /%					Speed /(frame·s^-1)	Model Size /MB
Network	AgeDB	CFP_FP	CFP_FF	LFW	CALFW	Speed /(frame·s^-1)	Model Size /MB
ResNet-101^[13]	97.28	95.11	99.65	99.71	96.65	42.64	250
ResNet-50^[13]	96.03	94.06	99.62	99.52	95.36	70.84	174.5
DenseNet-201(k=32)^[12]	96.68	94.83	99.62	99.68	96.04	100.17	161.8
DenseNet-169(k=32)^[12]	95.38	93.66	99.01	98.86	95.28	120.34	114.4
ShuffleNet(1×,g=3)^[22]	89.27	89.09	97.75	98.70	93.06	410.78	7.4
MobileNet-v1^[20]	88.65	88.54	97.06	98.43	93.01	206.64	13.7
MobileNet-v2^[21]	88.81	88.53	97.36	98.38	92.88	230.71	8.6
MobileFaceNet^[11]	92.95	89.46	98.03	98.96	93.89	432.41	4.1
Dual-MobileFaceNet	93.94	91.16	98.68	99.18	94.02	326.35	8.8

Table 2. Comparison of experiment results of different networks

Algorithm	LFW	CFP-FP	AgeDB-30
DeepFace^[3]	95.53	87.46	89.61
Deep FR^[23]	96.04	88.26	90.13
DeepID2^[4]	96.14	87.85	90.26
FaceNet^[5]	96.95	88.20	90.69
SphereFace^[6]	97.58	90.03	91.84
CosFace^[7]	98.43	90.75	92.33
InsightFace^[8]	99.18	91.16	93.94
O-Double classifier	99.12	91.21	93.22
Double classifier	99.46	93.33	95.88

Table 3. Recognition accuracy comparison of different algorithms%

Network	Recognition accuracy /%		Speed /(frame·s^-1)		FLOPS/10⁶
Network	8-people	18-people	8-people	18-people	FLOPS/10⁶
ResNet-101^[13]	97.08	94.14	2.16	4.37	22.69×10³
ResNet-50^[13]	95.96	91.51	1.28	2.61	12.34×10³
DenseNet-201^[12]	96.78	94.98	1.16	2.36	8.5×10³
DenseNet-169^[12]	95.27	91.69	0.89	1.81	6.6×10³
ShuffleNet^[22]	92.05	87.53	0.12	0.26	591
MobileNet-v1^[20]	91.12	85.60	0.16	0.35	1.1×10³
MobileNet-v2^[21]	91.96	86.33	0.13	0.28	1.0×10³
MobileFaceNet^[11]	92.83	88.77	0.10	0.21	439.8
Dual-MobileFaceNet	96.24	94.68	0.14	0.29	1.0×10³

Table 4. Experimental results of different networks on pan tilt video

Algorithm	8-people	18-people
DeepFace^[3]	87.53	83.67
Deep FR^[23]	88.54	84.27
DeepID2^[4]	88.94	84.25
FaceNet^[5]	89.35	85.33
SphereFace^[6]	90.58	87.68
CosFace^[7]	91.83	90.75
InsightFace^[8]	93.69	91.68
Double classifier	96.24	94.68

Table 5. Recognition accuracy of different algorithms%

Xichu Tian, Hansong Su, Gaohua Liu, Tengteng Liu. Improved Classroom Face Recognition Algorithm Based on InsightFace and Its Application[J]. Laser & Optoelectronics Progress, 2020, 57(22): 221501

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information