ComboLoss for Facial Attractiveness Analysis with Squeeze-and-Excitation Networks

Loss function is crucial for model training and feature representation learning, conventional models usually regard facial attractiveness recognition task as a regression problem, and adopt MSE loss or Huber variant loss as supervision to train a deep convolutional neural network (CNN) to predict facial attractiveness score. Little work has been done to systematically compare the performance of diverse loss functions. In this paper, we firstly systematically analyze model performance under diverse loss functions. Then a novel loss function named ComboLoss is proposed to guide the SEResNeXt50 network. The proposed method achieves state-of-the-art performance on SCUT-FBP, HotOrNot and SCUT-FBP5500 datasets with an improvement of 1.13%, 2.1% and 0.57% compared with prior arts, respectively. Code and models are available at this https URL.

[1]  Lu Xu,et al.  Hierarchical Multi-Task Network For Race, Gender and Facial Attractiveness Recognition , 2019, 2019 IEEE International Conference on Image Processing (ICIP).

[2]  Lianwen Jin,et al.  Attribute-Aware Convolutional Neural Networks for Facial Beauty Prediction , 2019, IJCAI.

[3]  Lianwen Jin,et al.  R2-ResNeXt: A ResNeXt-Based Regression Model with Relative Ranking for Facial Beauty Prediction , 2018, 2018 24th International Conference on Pattern Recognition (ICPR).

[4]  D. Perrett,et al.  Facial shape and judgements of female attractiveness , 1994, Nature.

[5]  Jie Xu,et al.  SCUT-FBP: A Benchmark Dataset for Facial Beauty Perception , 2015, 2015 IEEE International Conference on Systems, Man, and Cybernetics.

[6]  Anne Elorza Deias.  Face beauty analysis via manifold based semi-supervised learning , 2017 .

[7]  Jie Xu,et al.  Facial attractiveness prediction using psychologically inspired convolutional neural network (PI-CNN) , 2017, 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[8]  Quoc V. Le,et al.  EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks , 2019, ICML.

[9]  Xiaohui Yuan,et al.  CRNet: Classification and Regression Neural Network for Facial Beauty Prediction , 2018, PCM.

[10]  Lianwen Jin,et al.  Regression Guided by Relative Ranking Using Convolutional Neural Network (R$^3$3CNN) for Facial Beauty Prediction , 2019, IEEE Trans. Affect. Comput..

[11]  Zhuowen Tu,et al.  Aggregated Residual Transformations for Deep Neural Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  Bo Li,et al.  Facial attractiveness computation by label distribution learning with deep CNN and geometric features , 2017, 2017 IEEE International Conference on Multimedia and Expo (ICME).

[13]  Daniel Cohen-Or,et al.  A Humanlike Predictor of Facial Attractiveness , 2006, NIPS.

[14]  Xiaohui Yuan,et al.  Transferring Rich Deep Features for Facial Beauty Prediction , 2018, ArXiv.

[15]  Lianwen Jin,et al.  SCUT-FBP5500: A Diverse Benchmark Dataset for Multi-Paradigm Facial Beauty Prediction , 2018, 2018 24th International Conference on Pattern Recognition (ICPR).

[16]  Ming Shao,et al.  Attractive or Not?: Beauty Prediction with Attractiveness-Aware Encoders and Robust Late Fusion , 2014, ACM Multimedia.

[17]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[18]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  Eytan Ruppin,et al.  Facial Attractiveness: Beauty and the Machine , 2006, Neural Computation.

[20]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[21]  Yihong Gong,et al.  Predicting Facial Beauty without Landmarks , 2010, ECCV.

[22]  Ashok Samal,et al.  A landmark-based data-driven approach on 2.5D facial attractiveness computation , 2017, Neurocomputing.