.png){kind=icon}
Return
Face alignment using view-based direct appearance models
.png){kind=icon}
PRE
AI
DOI：10.1002/ima.10039.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Accurate face alignment is the prerequisite for many computer vision problems, such as face recognition, synthesis and 3D face modeling. In this article, a novel appearance model, called direct appearance model (DAM), is proposed and its extended view-based models are applied for multiview face alignment. Similar to the active appearance model (AAM), DAM also makes ingenious use of both shape and texture constraints; however, it does not combine them as in AAM; texture information is used directly to predict the shape and estimate the position and appearance (hence the name DAM). The way that DAM models shapes and textures has the following advantages as compared with AAM: (1) DAM subspaces include admissible appearances previously unseen in AAM, (2) it can converge more quickly and has higher accuracy, and (3) the memory requirement is cut down to a large extent. Extensive experiments are presented to evaluate the DAM alignment in comparison with AAM. (C) 2003 Wiley Periodicals, Inc.
Keywords:
direct appearance model
active appearance model
multiview face alignment
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
2.5
Papers:
2.1K
Citations:
2.3K
Organization
No organization information available
Cited Papers
Understanding the effect of GO on nitrogen assimilation in wheat through transcriptomics and metabolic process analysis.png)
Chemosphere
ChemosphereIF0
PREAI
ShareSaveShareSave
ShareSaveShareSave
View more