Automatic Selection of Key Points for 3D-Face Deformation

Tuan, Huynh Cao; Toàn, Đỗ Năng; Hien, Lam Thanh

doi:10.12720/jait.13.4.332-337

Cited by 4 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yang et al [22] propose FoldingNet which deforms a 2D grid into a 3D point cloud while preserving locality information. Tuan et al [23] proposed an innovative algorithm to automatically identify key points and cluster similar deformations with the radial basic function (RBF) technique to improve the deformation of 3D faces. Another study is concerned with the problem of synthesizing a face model from a set of basic surface models [24].…”

Section: Introductionmentioning

confidence: 99%

An algorithm for decomposing variations of 3D model

Phuong,

Hien,

Duc Vinh

et al. 2024

IJECE

View full text Add to dashboard Cite

In recent times, there has been an increasing number of people who are concerned about the virtual reality field. Parameterization of deformations of 3D models is a meaningful problem in theoretical research and application development of virtual reality. This paper proposes a technique for conditional decomposition of 3D model variations based on a given set of 3D observations of an object, along with a set of input strain weights. The proposed algorithm is conducted through an optimal iterative process with solving the non-negative least squares problem. The output of the technique is a set of base models corresponding to different types of strain. The result of the proposed technique allows the creation of a new 3D model variant of the object in a simple and visually observable way. The algorithm has been tested and proven effective on data that are 3D face models created from the Japanese Female Facial Expression (JAFFE) dataset with labeled expression weights.

show abstract

Section: Introductionmentioning

confidence: 99%