3DMM
A classic parametric representation of human faces proposed by Blanz and Vetter in 1999. It applies PCA dimensionality reduction to a large number of 3D face scans to obtain low-dimensional basis vectors, describing facial shape and texture variation with a small number of coefficients.
Timeline (last 90 days)
3DMM (3D Morphable Model) and its variants (such as FLAME and FaceVerse) use low-dimensional coefficients to separately control identity shape, expression deformation, and albedo, and the expression coefficients are usually obtained by regression from audio, video, or keypoints.
Linear parametric models (such as 3DMM) struggle to express high-frequency details such as wrinkles and dimples, so the mainstream approach is "parametric model as the base, neural representation for details".
Traditional avatar methods mostly rely on 3D morphable models (3DMM) to drive expressions, but 3DMM has limited expressive capability in regions such as the tongue and the interior of the mouth.
3DMM is a classic parametric face representation proposed by Blanz and Vetter in 1999, whose core idea is to apply PCA dimensionality reduction to a large number of 3D face scans.
All Facts (4)
3DMM (3D Morphable Model) and its variants (such as FLAME and FaceVerse) use low-dimensional coefficients to separately control identity shape, expression deformation, and albedo, and the expression coefficients are usually obtained by regression from audio, video, or keypoints.
50%UnverifiedLinear parametric models (such as 3DMM) struggle to express high-frequency details such as wrinkles and dimples, so the mainstream approach is "parametric model as the base, neural representation for details".
50%UnverifiedTraditional avatar methods mostly rely on 3D morphable models (3DMM) to drive expressions, but 3DMM has limited expressive capability in regions such as the tongue and the interior of the mouth.
50%Unverified3DMM is a classic parametric face representation proposed by Blanz and Vetter in 1999, whose core idea is to apply PCA dimensionality reduction to a large number of 3D face scans.
50%