Landmark Localization

A deformable object is commonly represented with a set of landmarks points which correspond to semantically meaningful parts. Landmark Localization is the problem of localizing these landmark points that correspond to a deformable object in an input image. This problem is commonly faced by training a deformable model of the object and try to fit it to the test image.

The methods that are currently implemented in menpofit are:

• Active Appearance Model (AAM)
  • Model Variants: Holistic, Patch-based, Masked, Linear, Linear Masked
  • Optimization Algorithms: Lucas-Kanade, Cascaded-Regression
• Active Pictorial Structures (APS)
  • Model Variant: Generative
  • Optimization Algorithm: Weighted Gauss-Newton Optimisation with fixed Jacobian and Hessian
• Constrained Local Model (CLM)
  • Active Shape Model
  • Regularised Landmark Mean Shift
• Ensemble of Regression Trees (ERT)
  • [provided by DLib]
• Supervised Descent Method (SDM)
  • Model Variants: Non Parametric, Parametric Shape, Parametric Appearance, Fully Parametric