Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 4, 2026World Electric Vehicle JournalOpen Access

Human Facial Keypoint Localization Based on T-Shaped Features and the Supervised Descent Method (TSDM)

View Full Paper
Ask AI
Bookmark
Share

Authors

YHYi-Wen HeXHXiao-ci Huang

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates enhanced facial landmark localization, suggesting practical applications in driver monitoring systems.

Key Points

  • To develop a novel method for accurate facial landmark localization using T-shaped features and the Supervised Descent Method.
  • Integrate T-shaped features with the Supervised Descent Method for localization.
  • Use AdaBoost for selecting T-shaped features to enhance face detection robustness.
  • Evaluate TSDM against traditional methods and lightweight deep learning models for performance.
  • TSDM achieves a face detection rate of 97.43% with a normalized mean error of 3.4%.
  • Demonstrates higher accuracy and lower false-positive rates than traditional methods like Haar and LBPH.
  • Outperforms several lightweight deep learning models in real-time performance on CPU-only platforms.

Cite This Study

He et al. (2026) studied this question.

synapsesocial.com/papers/69f837ab3ed186a739981dfehttps://doi.org/10.3390/wevj17050237
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Attention-Base deep learning for 3D craniofacial soft tissue landmark detection and diagnosis in orthodontics2025 · 3 citations
  2. 2FaceLift: Semi-supervised 3D Facial Landmark Localization2024
  3. 3Geometric Feature Enhancement for Robust Facial Landmark Detection in Makeup Paper Templates2026 · 1 citations
  4. 4Infinite 3D Landmarks: Improving Continuous 2D Facial Landmark Detection2024 · 5 citations
  5. 5A Robust and Efficient Method for Effective Facial Keypoint Detection2024 · 2 citations