마스크를 착용한 얼굴 인식을 위한 방법 연구
A Study for Face Recognition Wearing a Mask
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
현재 얼굴 인식 기술은 얼굴이 가려질 경우 인식률이 낮아지는 문제가 있다. 본 논문은 마스크를 착용한 얼굴의 인식률을 높이기 위해 얼굴 하단부에 모자이크를 씌워 이미지를 만든 후 유사도를 거리로 측정하여 높은 유사도를 보이는 두 얼굴을 매칭하는 방법을 제시한다. 제안한 방법은 마스크를 착용하지 않을 때와 유사한 인식률을 보이며 100명의 그룹에서 마스크를 착용한 경우 12% 향상된 Rank-1과 18.8% 향상된 매칭 정확도를 보인다.
Abstract
Recently, face recognition with occlusions has a problem in that the recognition rate is decreased. This paper proposes a method of matching two faces with high similarity by measuring the similarity by a distance after making an image with a mosaic on the lower part of the face to increase the recognition rate of a face wearing a mask. The proposed method shows a similar recognition rate as when the mask is not worn, and in the group of 100 people, the mask shows 12% improvement in Rank-1 and 18.8% improvement in matching accuracy.
참고문헌 (12)
- Ali, M. M., "Overview of fingerprint recognition system", International Conference on Electrical, Electronics, and Optimization Techniques, pp.1334~1338 (2016)
- Matin, A., "Human iris as a biometric for identity verification", 2nd International Conference on Electrical, Computer & Telecommunication Engineering, pp.1~4 (2016)
- Song, L., "Occlusion robust face recognition based on mask learning with pairwise differential siamese network", Proceedings of the IEEE International Conference on Computer Vision, pp.773~782 (2019)
- Redmon, J., "You Only Look Once : Unified, Real-Time Object Detection", IEEE Conference on Computer Vision and Pattern Recognition, pp.779~788 (2016)
- Liu, W., "Ssd : Single shot multibox detector", European conference on computer vision, pp.21~37 (2016)
- Chen, D., "Joint cascade face detection and alignment", European conference on computer vision, pp.109~122 (2014)
- Zhang, K., "Joint face detection and alignment using multitask cascaded convolutional networks", IEEE Signal Processing Letters, 23(10), pp.1499~1503 (2014)
- JDeng, J., "ArcFace : Additive Angular Margin Loss for Deep Face Recognition", IEEE Conference on Computer Vision and Pattern Recognition, pp.4690~4699 (2019)
- Deng, J., "RetinaFace : Single-Shot Multi-Level Face Localisation in the Wild", Proceedings of the IEEE, 2020, 5203~5212
- Yang, S., "Wider face : A face detection benchmark", IEEE conference on computer vision and pattern recognition, pp.5525~5533 (2016)
- He, K., "Deep residual learning for image recognition", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.770~778 (2016)
- Guo, Y., "Ms-celeb-1m : A dataset and benchmark for large-scale face recognition", European conference on computer vision, pp.87~102 (2016)