Patent US7426292B2 successfully refreshed from Google Patents.
← Back to Patent Explorer
US7426292B2 Expired / Lapsed US ⏱️ 0.0 years remaining

Method for determining optimal viewpoints for 3D face modeling and face recognition

Assignee: Mitsubishi Electric Research Laboratories Inc
•
Filing: 2004-04-30
•
Granted: 2008-09-16
•
Expires: 2024-04-30
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract A method determines an optimal set of viewpoints to acquire a 3D shape of a face. A view-sphere is tessellated with a plurality of viewpoint cells. The face is at an approximate center of the view-sphere. Selected viewpoint cells are discarded. The remaining viewpoint cells are clustered to a predetermined number of viewpoint cells according to a silhouette difference metric. The predetermined number of viewpoint cells are searched for a set of optimal viewpoint cells to construct a 3D model of the face.

Inventors

  • 👤 Baback Moghaddam
  • 👤 Hanspeter Pfister
  • 👤 Jinho Lee

Bibliographic Details

Application Number —
Jurisdiction US
Data Source google_patents
Last Synchronized 2026-11-09 12:19
CPC Classifications
G; G06; G06T; G06T17/00; G06V; G06V40/00; G06V40/10; G06V40/16; G06V40/172

Claims Analysis

Total Claims: 22
Claim 1
Independent Claim 89 words
A method for determining an optimal set of viewpoints to construct a 3D shape of a face from 2D images acquired at the optimal set of viewpoints, comprising: tessellating a view-sphere with a plurality of viewpoint cells, where the face is located at an approximate center of the view-sphere; discarding selected viewpoint cells; clustering remaining viewpoint cells to a predetermined number of viewpoint cells according to a silhouette difference metric determined from the 2D images; and searching the predetermined number of viewpoint cells for the optimal set of viewpoints.
Claim 2
Independent Claim 89 words
A method for determining an optimal set of viewpoints to construct a 3D shape of a face from 2D images acquired at the optimal set of viewpoints, comprising: tessellating a view-sphere with a plurality of viewpoint cells, where the face is located at an approximate center of the view-sphere; discarding selected viewpoint cells; clustering remaining viewpoint cells to a predetermined number of viewpoint cells according to a silhouette difference metric determined from the 2D images; and searching the predetermined number of viewpoint cells for the optimal set of viewpoints.
Claim 3
Dependent Claim 12 words
The method of claim 1 , in which the tessellation uses triangles.
Claim 4
Dependent Claim 17 words
The method of claim 1 , in which the tessellation is a uniform subdivision of a dodecahedron.
Claim 5
Dependent Claim 21 words
The method of claim 1 , in which the selected viewpoint cells include viewpoint cells in a rear-hemisphere of the view-sphere.
Claim 6
Dependent Claim 21 words
The method of claim 1 , in which the selected viewpoint cells include viewpoint cells above and below a predetermined elevation.
Claim 7
Dependent Claim 21 words
The method of claim 5 , in which the predetermined elevation is ±45° from a central horizontal plane of the view-sphere.
Claim 8
Dependent Claim 32 words
The method of claim 1 , in which the selected viewpoint cells include viewpoint cells of half of the face due to an approximate bilateral symmetry of two halves of the face.
Claim 9
Dependent Claim 20 words
The method of claim 1 , in which the silhouette difference metric is measured for each pair of viewpoint cells.
Claim 10
Dependent Claim 16 words
The method of claim 1 , in which the predetermined number of viewpoint cells is ten.
Claim 11
Dependent Claim 25 words
The method of claim 1 , in which a location of the viewpoint cells is determined by a centroid of the cluster of viewpoint cells.
Claim 12
Dependent Claim 18 words
The method of claim 1 , in which only the silhouette differences near critical facial areas are considered.
Claim 13
Dependent Claim 22 words
The method of claim 11 , in which the critical facial areas include a nose, eyes, ears, a chin, and a mouth.
Claim 14
Dependent Claim 18 words
The method of claim 1 , in which the silhouette distances are stored in a precomputed look-up table.
Claim 15
Dependent Claim 12 words
The method of claim 1 , in which the searching is exhaustive.
Claim 16
Dependent Claim 32 words
The method of claim 1 , in which the set of optimal viewpoint cells is applied to a sequence of images in a video to select automatically optimal poses for face modeling.
Claim 17
Dependent Claim 32 words
The method of claim 1 , in which the set of optimal viewpoint cells is applied to a sequence of images in a video to select automatically optimal poses for face recognition.
Claim 18
Dependent Claim 34 words
The method of claim 1 , in which the set of optimal viewpoint cells is used to construct the 3D model of the face from 2D images acquired at the optimal set of viewpoints.
Claim 19
Dependent Claim 25 words
The method of claim 1 , in which the searching is a combinatorial search of all possible subsets of the optimal set of viewpoint cells.
Claim 20
Dependent Claim 27 words
The method of claim 1 , in which optimality is defined by a minimum reconstruction error between an input face and the 3D model of the face.
Claim 21
Dependent Claim 21 words
The method of claim 19 , in which the minimum reconstruction error is based on a combination of shape and texture.
Claim 22
Dependent Claim 22 words
The method of claim 19 , in which the minimum reconstruction error is based on an average of a plurality of faces.

References Cited (Backward Citations)

US5524396A Cited
US6141034A Cited
US5748199A Cited
US5991437A Cited
US6879324B1 Cited
US6792398B1 Cited
US6831640B2 Cited
US6307555B1 Cited
US6152739A Cited
US6807286B1 Cited

USPTO Annuity & Maintenance Windows

Annuity Stage Window Opens Payment Deadline Surcharge Period Until Estimated Fee Status
3.5 Years (Year 4) 2011-09-16 2012-03-17 2012-09-16 $2000.00 Window Closed / Lapsed
7.5 Years (Year 8) 2015-09-16 2016-03-17 2016-09-16 $3760.00 Window Closed / Lapsed
11.5 Years (Year 12) 2019-09-16 2020-03-17 2020-09-16 $7700.00 Window Closed / Lapsed

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag