Bidirectional Warping of Active Appearance Model
This work addresses a specific bottleneck in facial image analysis for applications like face identification and expression recognition, representing an incremental improvement over existing AAM fitting methods.
The paper tackled the problem of improving Active Appearance Model (AAM) fitting for facial image analysis by proposing a bidirectional warping approach that warps both the input image and appearance template, resulting in outperformance over state-of-the-art inverse compositional methods on the Multi-PIE database for landmark extraction with shape and pose variations.
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.