《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (4): 1300-1308.DOI: 10.11772/j.issn.1001-9081.2025040398

• 多媒体计算与计算机仿真 • 上一篇    

基于动态卷积自编码器的无监督人脸属性编辑方法

崔选1,2, 刘波1,2()   

  1. 1.重庆工商大学 人工智能学院,重庆 400072
    2.智能感知与区块链技术重庆市重点实验室(重庆工商大学),重庆 400072
  • 收稿日期:2025-04-14 修回日期:2025-06-30 接受日期:2025-07-02 发布日期:2025-07-07 出版日期:2026-04-10
  • 通讯作者: 刘波
  • 作者简介:崔选(1999—),男,湖南娄底人,硕士研究生,主要研究方向:计算机视觉、图像处理
  • 基金资助:
    2023年中国高校产学研创新基金资助项目(2023ZY024);重庆工商大学研究生创新项目(yjscxx2025-269-181)

Unsupervised face attribute editing method based on dynamic convolutional autoencoder

Xuan CUI1,2, Bo LIU1,2()   

  1. 1.College of Artificial Intelligence,Chongqing Technology and Business University,Chongqing 400072,China
    2.Chongqing Key Laboratory of Intelligent Perception and Blockchain Technology (Chongqing Technology and Business University),Chongqing 400072,China
  • Received:2025-04-14 Revised:2025-06-30 Accepted:2025-07-02 Online:2025-07-07 Published:2026-04-10
  • Contact: Bo LIU
  • About author:CUI Xuan, born in 1999, M. S. candidate. His research interests include computer vision, image processing.
  • Supported by:
    2023 China University Industry-University-Research Innovation Fund(2023ZY024);Graduate Innovation Project of Chongqing Technology and Business University(yjscxx2025-269-181)

摘要:

基于生成对抗网络(GAN)潜空间的无监督人脸属性编辑方法具有效率高和无需标注数据的优点,然而这类方法在解耦性和可控性方面仍面临挑战,如在操控特定人脸属性时,可能会引起其他属性的意外变化,从而影响编辑效果,以及难以精确控制所编辑人脸属性的变化程度。针对上述问题,提出基于动态卷积自编码器的无监督人脸属性编辑(AUFAE)方法。该方法通过在潜空间中学习有效的语义向量,实现对人脸属性的精准编辑。具体地,设计动态卷积自编码器网络(DCAE-Net)作为主干网络,该网络的编码器部分采用动态卷积(DyConv)的方式动态提取潜空间的局部特征,从而学习具有局部特性的语义向量;在解码器部分则融入通道注意力(CA)机制建立通道间的非线性依赖关系,使模型能够自主地聚焦不同语义相关的特征通道,有效促进语义向量的独立性学习。为了增强语义向量的解耦性和可控性,引入基于属性边界向量的损失函数训练DCAE-Net。此外,引入软正交损失确保语义向量之间相互独立,进一步提升解耦性能。实验结果表明,在3个预训练GAN生成模型上,与3种主流的人脸属性编辑方法相比,AUFAE的弗雷歇距离(FID)减小了37.43%~50.21%,学习感知图像块相似度(LPIPS)减小了23.61%~42.85%,结构相似性指数(SSIM)提高了7.04%~13.42%。在直观视觉上,AUFAE在人脸属性编辑过程中也未出现属性耦合现象。可见,AUFAE能够有效地缓解人脸编辑过程中的属性耦合现象,并实现更精确的人脸属性编辑。

关键词: 生成对抗网络, 语义向量, 人脸属性编辑, 属性边界向量, 动态卷积

Abstract:

Unsupervised face attribute editing methods based on the latent space of Generative Adversarial Networks (GANs) offer advantages of high efficiency and only label-free data required, but they still face challenges in terms of decoupling and controllability, for instance, modifying a specific face attribute may alter other attributes inadvertently, thereby affecting editing quality, and precise control over the degree of attribute modification remains difficult. To address these issues, a dynamic convolutional Autoencoder-based Unsupervised Face Attribute Editing (AUFAE) method was proposed to achieve precise face attribute editing by learning effective semantic vectors in the latent space. Specifically, a Dynamic Convolutional AutoEncoder Network (DCAE-Net) was designed as the backbone, where Dynamic Convolution (DyConv) was utilized by the encoder to extract local latent-space features adaptively, thereby learning semantic vectors with local characteristics. A Channel Attention (CA) mechanism was incorporated into the decoder to establish nonlinear dependencies between channels, thereby allowing the model to focus on feature channels relevant to different semantics autonomously and enhancing the independence of semantic vector learning. To improve decoupling and controllability of semantic vectors, an attribute boundary vector-based loss function was introduced to train DCAE-Net. Additionally, a soft orthogonality loss was applied to ensure mutual independence of semantic vectors, thereby further boosting decoupling performance. Experimental results show that on three pre-trained GAN generation models, compared with three mainstream face attribute editing methods, AUFAE has the Fréchet Inception Distance (FID) decreased by 37.43%-50.21%, the Learned Perceptual Image Patch Similarity (LPIPS) decreased by 23.61%-42.85%, and the Structural Similarity Index Measure (SSIM) increased by 7.04%-13.42%. On intuitive vision, AUFAE does not exhibit attribute coupling during face attribute editing process. It can be seen that AUFAE can alleviate the attribute coupling in face editing process effectively and achieve more accurate face attribute editing.

Key words: Generative Adversarial Network (GAN), semantic vector, face attribute editing, attribute boundary vector, Dynamic Convolution (DyConv)

中图分类号: