KeyPoint Relative Position Encoding for Face Recognition

arXiv:2403.14852v1 Announce Type: new
Abstract: In this paper, we address the challenge of making ViT models more robust to unseen affine transformations. Such robustness becomes useful in various recognition tasks such as face recognition when image alignment failures occur. We propose a novel method called KP-RPE, which leverages key points (e.g.~facial landmarks) to make ViT more resilient to scale, translation, and pose variations. We begin with the observation that Relative Position Encoding (RPE) is a good way to bring affine …

abstract alignment arxiv challenge cs.cv encoding face face recognition image key making novel paper recognition robust robustness tasks type vit