Proto-Former: Unified Facial Landmark Detection by Prototype Transformer

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hu, Shengkai, Qi, Haozhe, Wan, Jun, Huang, Jiaxing, Zhang, Lefei, Sun, Hang, Tao, Dacheng
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909852564979712
author Hu, Shengkai
Qi, Haozhe
Wan, Jun
Huang, Jiaxing
Zhang, Lefei
Sun, Hang
Tao, Dacheng
author_facet Hu, Shengkai
Qi, Haozhe
Wan, Jun
Huang, Jiaxing
Zhang, Lefei
Sun, Hang
Tao, Dacheng
contents Recent advances in deep learning have significantly improved facial landmark detection. However, existing facial landmark detection datasets often define different numbers of landmarks, and most mainstream methods can only be trained on a single dataset. This limits the model generalization to different datasets and hinders the development of a unified model. To address this issue, we propose Proto-Former, a unified, adaptive, end-to-end facial landmark detection framework that explicitly enhances dataset-specific facial structural representations (i.e., prototype). Proto-Former overcomes the limitations of single-dataset training by enabling joint training across multiple datasets within a unified architecture. Specifically, Proto-Former comprises two key components: an Adaptive Prototype-Aware Encoder (APAE) that performs adaptive feature extraction and learns prototype representations, and a Progressive Prototype-Aware Decoder (PPAD) that refines these prototypes to generate prompts that guide the model's attention to key facial regions. Furthermore, we introduce a novel Prototype-Aware (PA) loss, which achieves optimal path finding by constraining the selection weights of prototype experts. This loss function effectively resolves the problem of prototype expert addressing instability during multi-dataset training, alleviates gradient conflicts, and enables the extraction of more accurate facial structure features. Extensive experiments on widely used benchmark datasets demonstrate that our Proto-Former achieves superior performance compared to existing state-of-the-art methods. The code is publicly available at: https://github.com/Husk021118/Proto-Former.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15338
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Proto-Former: Unified Facial Landmark Detection by Prototype Transformer
Hu, Shengkai
Qi, Haozhe
Wan, Jun
Huang, Jiaxing
Zhang, Lefei
Sun, Hang
Tao, Dacheng
Computer Vision and Pattern Recognition
Recent advances in deep learning have significantly improved facial landmark detection. However, existing facial landmark detection datasets often define different numbers of landmarks, and most mainstream methods can only be trained on a single dataset. This limits the model generalization to different datasets and hinders the development of a unified model. To address this issue, we propose Proto-Former, a unified, adaptive, end-to-end facial landmark detection framework that explicitly enhances dataset-specific facial structural representations (i.e., prototype). Proto-Former overcomes the limitations of single-dataset training by enabling joint training across multiple datasets within a unified architecture. Specifically, Proto-Former comprises two key components: an Adaptive Prototype-Aware Encoder (APAE) that performs adaptive feature extraction and learns prototype representations, and a Progressive Prototype-Aware Decoder (PPAD) that refines these prototypes to generate prompts that guide the model's attention to key facial regions. Furthermore, we introduce a novel Prototype-Aware (PA) loss, which achieves optimal path finding by constraining the selection weights of prototype experts. This loss function effectively resolves the problem of prototype expert addressing instability during multi-dataset training, alleviates gradient conflicts, and enables the extraction of more accurate facial structure features. Extensive experiments on widely used benchmark datasets demonstrate that our Proto-Former achieves superior performance compared to existing state-of-the-art methods. The code is publicly available at: https://github.com/Husk021118/Proto-Former.
title Proto-Former: Unified Facial Landmark Detection by Prototype Transformer
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.15338