Linear Digressions Better Facial Recognition with Fisherfaces
5 snips
Jan 7, 2015 The podcast explores the challenges of facial recognition and how it can be affected by variations in expressions, lighting, and angle. The Fisherfaces algorithm is introduced as a more robust alternative to eigenfaces, utilizing a fisher linear discriminant to distinguish based on the smallest inter-class distance. The chapter descriptions discuss the exploration and analysis of facial features, the comparison between Fisher Linear Discriminate and Eigenfaces, and the superior performance of Fisher Faces in challenging situations.
AI Snips
Chapters
Transcript
Episode notes
Mugshot Training Bias
- Training on uniform mugshots trains models to focus on incidental traits like lighting and expression.
- Models may misidentify different people who share those incidental traits.
PCA Prioritizes Variance Over Identity
- PCA (eigenfaces) finds directions of maximal variance, not identity-defining features.
- This lets lighting or background dominate recognition instead of who the person is.
Fisher LDA Focuses On Class Separation
- Fisher Linear Discriminant projects data to maximize between-class distance and minimize within-class spread.
- This creates tight class clusters that are easier to separate for classification.