Finding Optimal Views for 3D Face Shape Modeling

DSpace/Manakin Repository

Finding Optimal Views for 3D Face Shape Modeling

Citable link to this page

 

 
Title: Finding Optimal Views for 3D Face Shape Modeling
Author: Lee, Jinho; Moghaddam, Baback; Pfister, Hanspeter; Machiraju, Raghu

Note: Order does not necessarily reflect citation order of authors.

Citation: Lee, Jinho, Baback Moghaddam, Hanspeter Pfister, and Raghu Machiraju. 2004. Finding optimal views for 3D face shape modeling. In Proceedings Sixth IEEE International Conference on Automatic Face and Gesture Recognition: FGR 2004: May 17-19, 2004, Seoul, Korea, ed. K. Taehakkyo, 31-36. Los Alamitos, C.A.: IEEE Computer Society.
Full Text & Related Files:
Abstract: A fundamental problem in multi-view 3D face modeling is the determination of the set of optimal views (poses) required for accurate 3D shape estimation of a generic face. There is no analytical solution to this problem, instead (partial) solutions require (near) exhaustive combinatorial search, hence the inherent computational difficulty of this task. We build on our previous modeling framework [Silhouette-based 3D face shape recovery, Model-based 3D face capture using shape-from-silhouettes] which uses an efficient contour-based silhouette method and extend it by aggressive pruning of the view-sphere with view clustering and various imaging constraints. A multi-view optimization search is performed using both model-based (eigenheads) and data-driven (visual hull) methods, yielding comparable best views. These constitute the first reported set of optimal views for 3D face shape capture and provide useful empirical guidelines for the design of 3D face recognition systems.
Published Version: doi:10.1109/AFGR.2004.1301505
Other Sources: http://gvi.seas.harvard.edu/sites/all/files/fg04_0.pdf
Terms of Use: This article is made available under the terms and conditions applicable to Other Posted Material, as set forth at http://nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of-use#LAA
Citable link to this page: http://nrs.harvard.edu/urn-3:HUL.InstRepos:4138674
Downloads of this work:

Show full Dublin Core record

This item appears in the following Collection(s)

 
 

Search DASH


Advanced Search
 
 

Submitters