DOI QR코드

DOI QR Code

Facial Age Classification and Synthesis using Feature Decomposition

특징 분해를 이용한 얼굴 나이 분류 및 합성

  • Chanho Kim (Inha University, Department of Electrical and Computer Engineering) ;
  • In Kyu Park (Inha University, Department of Electrical and Computer Engineering)
  • 김찬호 (인하대학교 전기컴퓨터공학과) ;
  • 박인규 (인하대학교 전기컴퓨터공학과)
  • Received : 2023.02.20
  • Accepted : 2023.03.17
  • Published : 2023.03.30

Abstract

Recently deep learning models are widely used for various tasks such as facial recognition and face editing. Their training process often involves a dataset with imbalanced age distribution. It is because some age groups (teenagers and middle age) are more socially active and tends to have more data compared to the less socially active age groups (children and elderly). This imbalanced age distribution may negatively impact the deep learning training process or the model performance when tested against those age groups with less data. To this end, we propose an age-controllable face synthesis technique using a feature decomposition to classify age from facial images which can be utilized to synthesize novel data to balance out the age distribution. We perform extensive qualitative and quantitative evaluation on our proposed technique using the FFHQ dataset and we show that our method has better performance than existing method.

최근 인공지능 모델을 이용한 얼굴인식, 얼굴 수정 등 다양한 얼굴 작업들이 실생활에도 광범위하게 사용되고 있다. 그러나 모델의 학습에 사용되는 대부분의 얼굴 데이터셋은 사회활동이 활발한 특정 나이에 편중되고, 어린아이나 노인의 데이터가 적은 경향이 있다. 이와 같은 데이터셋 불균형 문제는 모델의 학습에도 좋지 않은 영향을 끼쳐, 아이나 노인같이 데이터가 적은 나이의 사람이 인공지능 모델을 사용할 때 사회활동이 활발한 나이의 사람이 사용할 때보다 성능이 떨어질 수 있고, 이들의 인공지능 모델 사용을 어렵게 할 가능성이 높다. 이를 개선하기 위해 본 논문은 특징 분해를 활용하여 얼굴 영상으로부터 나이를 분류하고 목표 나이로 합성하는 기법을 제안한다. 제안하는 기법은 FFHQ-Aging 데이터셋을 이용한 정량적, 정성적 평가를 통해 기존의 방법보다 더 나은 성능을 보인다.

Keywords

Acknowledgement

This work was partly supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No.2022R1A4A1033549 and No.NRF-2019R1A2C1006706). This work was partly supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (No.RS-2022-00155915, Artificial Intelligence Convergence Innovation Human Resources Development (Inha University)).

References

  1. X. Yao, G. Puy, A. Newson, Y. Gousseau, P. Hellier, "High resolution face age editing," Proc. International Conference on Pattern Recognition, January 2021. doi: https://doi.org/10.1109/icpr48806.2021.9412383
  2. F. Makhmudkhujaev, S. Hong, and I. K. Park, "Re-Aging GAN: Toward personalized face age transformation," Proc. IEEE/CVF International Conference on Computer Vision, October 2021. doi: https://doi.org/10.1109/iccv48922.2021.00388
  3. Z. Huang, J. Zhang, and H. Shan, "When age-invariant face recognition meets face age synthesis: A multi-task learning framework," Proc. IEEE/CVF Computer Vision and Pattern Recognition, June 2021. doi: https://doi.org/10.1109/cvpr46437.2021.00720
  4. R. Or-El, S. Sengupta, O. Fried, E. Shechtman, I. Kemelmacher-Shlizerman, "Lifespan age transformation synthesis," Proc. European Conference on Computer Vision, August 2020. doi: https://doi.org/10.1007/978-3-030-58539-6_44