Facial Recognition Software Use on Surgically Altered Faces: A Systematic Review.

Area of Science:

• Biometrics and Computer Vision
• Medical Imaging and Surgical Interventions
• Human-Computer Interaction

Background:

• Facial recognition software (FRS) is widely used in commercial, security, and entertainment applications.
• Surgical alterations to the face present a significant challenge to the accuracy and reliability of FRS algorithms.
• This study systematically reviews current methodologies for recognizing surgically modified faces.

Approach:

• A systematic literature review was conducted using PubMed and IEEE databases.
• 178 manuscripts related to FRS and surgery were initially identified.
• 39 relevant English-language studies focusing on FRS of surgically altered faces were included.

Key Points:

• Surgical procedures reviewed include skin treatments, rhinoplasty, facelifts, and facial feminization.
• Recognition methods were categorized into appearance-based, feature-based, and texture-based approaches.
• Reported accuracy rates for FRS on surgically altered faces ranged from 19.1% to 85.35%.

Conclusions:

• Knowledge of FRS limitations post-surgery can guide patient counseling on technology use and security.
• Understanding surgical impacts on facial features can enhance the development of more robust FRS algorithms.
• Improved FRS performance for postsurgical recognition is achievable through integrated surgical knowledge.