Bidirectional Warping of Active Appearance Model

TL;DR

This paper introduces a bidirectional warping method for Active Appearance Models that improves facial landmark detection accuracy by simultaneously warping both input images and appearance templates, outperforming previous methods.

Contribution

The paper presents a novel bidirectional warping approach for AAM fitting that combines affine updates and inverse compositional methods for better facial analysis.

Findings

Outperforms state-of-the-art inverse compositional approaches

Effective in handling shape and pose variations

Improves landmark detection accuracy

Abstract

Active Appearance Model (AAM) is a commonly used method for facial image analysis with applications in face identification and facial expression recognition. This paper proposes a new approach based on image alignment for AAM fitting called bidirectional warping. Previous approaches warp either the input image or the appearance template. We propose to warp both the input image, using incremental update by an affine transformation, and the appearance template, using an inverse compositional approach. Our experimental results on Multi-PIE face database show that the bidirectional approach outperforms state-of-the-art inverse compositional fitting approaches in extracting landmark points of faces with shape and pose variations.