Image-Driven Panel Design via Feature-Preserving Mesh Deformation - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2011

Image-Driven Panel Design via Feature-Preserving Mesh Deformation

Baojun Li
  • Function : Author
  • PersonId : 1012766
Xiuping Liu
  • Function : Author
  • PersonId : 1012767
Ping Hu
  • Function : Author
  • PersonId : 1012768


In this paper, we propose an image-driven 3D modeling technique for rapid panel design. Our semi-automatic approach is based on template technique and mesh volume deformation controlled by a special cage. We designed our modeling system to be interactive in 2D, automating the process of shape generation while relying on the user to provide image samples. Once a parametric model template is given, using the contour extracted from images, the new control cage corresponding to mesh models generated. Then the geometry of new panel is automatically recovered from the deformable template model. Our system also allows the user to easily reconstruct other 3D objects in a similar manner, such as realistic-looking plant modeling from images. We show realistic reconstructions of a variety of panels, automobile shapes and demonstrate examples of plant editing.
Fichier principal
Vignette du fichier
978-3-642-18336-2_5_Chapter.pdf (2.71 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01562795 , version 1 (17-07-2017)





Baojun Li, Xiuping Liu, Yanqi Liu, Ping Hu, Mingzeng Liu, et al.. Image-Driven Panel Design via Feature-Preserving Mesh Deformation. 4th Conference on Computer and Computing Technologies in Agriculture (CCTA), Oct 2010, Nanchang, China. pp.30-40, ⟨10.1007/978-3-642-18336-2_5⟩. ⟨hal-01562795⟩
48 View
36 Download



Gmail Facebook Twitter LinkedIn More