This user guide is a general introduction to MenpoFit, aiming to provide a bird's eye of MenpoFit's design. After reading this guide you should be able to go explore MenpoFit's extensive Notebooks and not be too surprised by what you see.
The vast majority of existing deformable modeling software suffers from one or more of the following important issues:
- It is released in binary closed-source format
- It does not come with training code; only pre-trained models
- It is not well-structured which makes it very difficult to tweak and alter
- It only focuses on a single method/model
MenpoFit overcomes the above issues by providing open-source training and fitting code for multiple state-of-the-art deformable models under a unified protocol. We strongly believe that this is the only way towards reproducable and high-quality research.
MenpoFit is an object oriented framework for building and fitting deformable models. It makes some basic assumptions that are common for all the implemented methods. For example, all deformable models are trained in multiple scales and the fitting procedure is, in most cases, iterative. MenpoFit's key interfaces are:
MultiScaleNonParametricFitter- multi-scale fitting classMultiScaleParametricFitter- multi-scale fitting class that uses a parametric shape modelMultiScaleNonParametricIterativeResult- multi-scale result of an iterative fittingMultiScaleParametricIterativeResult- multi-scale result of an iterative fitting using a parametric shape model
AAM,LucasKanadeAAMFitter,SupervisedDescentAAMFitter- Active Appearance Model builder and fittersATM,LucasKanadeATMFitter- Active Template Model builder and fitterGenerativeAPS,GaussNewtonAPSFitter- Active Pictorial Structures builder and fitterCLM,GradientDescentCLMFitter- Constrained Local Model builder and fitterLucasKanadeFitter- Lucas-Kanade Image AlignmentSupervisedDescentFitter- Supervised Descent Method builder and fitterDlibERT- Ensemble of Regression Trees builder and fitter