Pose-invariant face recognition via feature-space pose frontalization

2025-05-22

AI Summary

To address the challenging problem of pose-invariant face recognition—specifically, matching unconstrained profile-face images against frontal-face gallery images—this paper proposes a Feature Space Pose Frontalization Module (FSPFM) and an attention-guided two-stage training paradigm. Instead of pixel-level reconstruction—which often introduces artifacts—the method directly optimizes pose correction and identity discrimination jointly in the deep feature space. Leveraging a pre-trained feature encoder, it incorporates pose-aware feature alignment and attention-driven fine-tuning to simultaneously enhance pose robustness and identity discriminability. The approach achieves state-of-the-art performance on five major benchmarks: LFW, CFP-FP, AgeDB-30, CALFW, and CPFW. Moreover, it maintains strong generalization across standard face recognition tasks, demonstrating both effectiveness and broad applicability.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Feature Construction/Reformulation
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Large-scale security measurements

Abstract

Pose-invariant face recognition has become a challenging problem for modern AI-based face recognition systems. It aims at matching a profile face captured in the wild with a frontal face registered in a database. Existing methods perform face frontalization via either generative models or learning a pose robust feature representation. In this paper, a new method is presented to perform face frontalization and recognition within the feature space. First, a novel feature space pose frontalization module (FSPFM) is proposed to transform profile images with arbitrary angles into frontal counterparts. Second, a new training paradigm is proposed to maximize the potential of FSPFM and boost its performance. The latter consists of a pre-training and an attention-guided fine-tuning stage. Moreover, extensive experiments have been conducted on five popular face recognition benchmarks. Results show that not only our method outperforms the state-of-the-art in the pose-invariant face recognition task but also maintains superior performance in other standard scenarios.

Problem

• Pose-invariant face recognition for profile-to-frontal matching
• Feature-space pose frontalization to transform profile images
• Outperforming state-of-the-art in pose-invariant recognition tasks

Innovation

• Feature-space pose frontalization module (FSPFM) transforms profile images
• Training paradigm includes pre-training and fine-tuning stages
• Outperforms state-of-the-art in pose-invariant face recognition