U-Face: An Efficient and Generalizable Framework for Unsupervised Facial Attribute Editing via Subspace Learning

TL;DR

U-Face introduces a novel subspace learning framework for unsupervised facial attribute editing that enhances disentanglement and controllability by imposing orthogonal constraints and using an iterative algorithm.

Contribution

The paper proposes a new subspace learning approach with orthogonal constraints and an iterative algorithm for improved unsupervised facial attribute editing.

Findings

Effective disentanglement of facial attributes achieved.

Enhanced controllability in attribute editing demonstrated.

Framework supports flexible and generalizable attribute manipulation.

Abstract

Latent space-based facial attribute editing methods have gained popularity in applications such as digital entertainment, virtual avatar creation, and human-computer interaction systems due to their potential for efficient and flexible attribute manipulation, particularly for continuous edits. Among these, unsupervised latent space-based methods, which discover effective semantic vectors without relying on labeled data, have attracted considerable attention in the research community. However, existing methods still encounter difficulties in disentanglement, as manipulating a specific facial attribute may unintentionally affect other attributes, complicating fine-grained controllability. To address these challenges, we propose a novel framework designed to offer an effective and adaptable solution for unsupervised facial attribute editing, called Unsupervised Facial Attribute Controllable Editing (U-Face). The proposed method frames semantic vector learning as a subspace learning problem, where latent vectors are approximated within a lower-dimensional semantic subspace spanned by a semantic vector matrix. This formulation can also be equivalently interpreted from a projection-reconstruction perspective and further generalized into an autoencoder framework, providing a foundation that can support disentangled representation learning in a flexible manner. To improve disentanglement and controllability, we impose orthogonal non-negative constraints on the semantic vectors and incorporate attribute boundary vectors to reduce entanglement in the learned directions. Although these constraints make the optimization problem challenging, we design an alternating iterative algorithm, called Alternating Iterative Disentanglement and Controllability (AIDC), with closed-form updates and provable convergence under specific conditions.