CLIP-FTI: Fine-Grained Face Template Inversion via CLIP-Driven Attribute Conditioning
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (5) · 3479-3487
DOI: 10.1609/aaai.v40i5.37345 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Longchen Dai und 4 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).
Abstract (Originalsprache)
Face recognition systems store face templates for efficient matching. Once leaked, these templates pose a threat: inverting them can yield photorealistic surrogates that compromise privacy and enable impersonation. Although existing research has achieved relatively realistic face template inversion, the reconstructed facial images exhibit over-smoothed facial-part attributes (eyes, nose, mouth) and limited transferability. To address this problem, we present CLIP-FTI, a CLIP-driven fine-grained attribute conditioning framework for face template inversion. Our core idea is to use the CLIP model to obtain the semantic embeddings of facial features, in order to realize the reconstruction of specific facial feature attributes. Specifically, facial feature attribute embeddings extracted from CLIP are fused with the leaked template via a cross-modal feature interaction network and projected into the intermediate latent space of a pretrained Style- GAN. The StyleGAN generator then synthesizes face images with the same identity as the templates but with more finegrained facial feature attributes. Experiments across multiple face recognition backbones and datasets show that our reconstructions (i) achieve higher identification accuracy and attribute similarity, (ii) recover sharper component-level attribute semantics, and (iii) improve cross-model attack transferability compared to prior reconstruction attacks. To the best of our knowledge, ours is the first method to use additional information besides the face template attack to realize face template inversion and obtains SOTA results.
Zitieren
APA
Dai, L., Shen, Z., Zhou, Z., Yu, P., & Xia, Z. (2026). CLIP-FTI: Fine-Grained Face Template Inversion via CLIP-Driven Attribute Conditioning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(5), 3479-3487. https://doi.org/10.1609/aaai.v40i5.37345
Vancouver
Dai L, Shen Z, Zhou Z, Yu P, Xia Z. CLIP-FTI: Fine-Grained Face Template Inversion via CLIP-Driven Attribute Conditioning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(5):3479-3487. doi:10.1609/aaai.v40i5.37345
BibTeX
@article{Dai2026,
author = {Dai, Longchen and Shen, Zixuan and Zhou, Zhiheng and Yu, Peipeng and Xia, Zhihua},
title = {CLIP-FTI: Fine-Grained Face Template Inversion via CLIP-Driven Attribute Conditioning},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {5},
pages = {3479--3487},
doi = {10.1609/aaai.v40i5.37345}
}
EUVIMED