A hybrid approach to building face shape classifier for hairstyle recommender system
A hybrid approach to building face shape classifier for hairstyle recommender system
复制标题
DOI：
10.1016/j.eswa.2018.11.011
复制
发表时间：
2019-04
期刊：
影响因子：
--
通讯作者：
Kitsuchart Pasupa;Wisuwat Sunhem;C. Loo
中科院分区：
文献类型：
--
作者：
Kitsuchart Pasupa;Wisuwat Sunhem;C. Loo
根据发型专家的指导，识别人脸形状是选择正确发型之前的第一个也是最重要的过程，尤其是对于女性。本文提出了一种基于人脸形状分类器的发型推荐系统框架。该框架实现了具有单个面部图像的自动发型推荐。这对美容行业的服务提供商产生了直接影响。它可以模拟当用户穿着专家系统推荐的选择发型时的样子。在这个框架中使用的模型是基于支持向量机。该框架在手工制作的，深度学习的（VGG-face）特征和VGG-face微调版本上进行评估，用于人脸形状分类任务。除了通过一个精心设计的框架来评估这些单独的特征之外，我们还试图将这三个描述符融合在一起，以提高分类任务的性能。采用了两种组合技术，即：向量级联和多核学习（MKL）技术。采用粒子群优化算法对模型的所有超参数进行了优化。结果表明，将手工制作的和VGG面部描述符与MKL相结合产生了最好的结果，准确率为70.3%，在统计学上显著优于使用单个特征。因此，将数据的多种表示与MKL相结合可以提高专家系统的整体性能。此外，这证明了手工制作的描述符可以与深度学习的描述符互补。
Identifying human face shape is the first and the most vital process prior to choosing the right hairstyle to wear on according to guidelines from hairstyle experts, especially for women. This work presents a novel framework for a hairstyle recommender system that is based on face shape classifier. This framework enables an automatic hairstyle recommendation with a single face image. This has a direct impact on beauty industry service providers. It can simulate how the user looks like when she is wearing the chosen hairstyle recommended by the expert system. The model used in this framework is based on Support Vector Machine. The framework is evaluated on hand-crafted, deep-learned (VGG-face) features and VGG-face fine-tuned version for the face shape classification task. In addition to evaluating these individual features by a well-designed framework, we attempted to fuse these three descriptors together in order to improve the performance of the classification task. Two combination techniques were employed, namely: Vector Concatenation and Multiple Kernel Learning (MKL) techniques. All the hyper-parameters of the model were optimised by using Particle Swarm Optimisation. The results show that combining hand-crafted and VGG-face descriptors with MKL yielded the best results at 70.3% of accuracy which was statistically significantly better than using individual features. Thus, combining multiple representations of the data with MKL can improve the overall performance of the expert system. In addition, this proves that hand-crafted descriptor can be complementary to deep-learned descriptor.