Download now Free registration required
In this paper, the authors explore the use of methodologies for 3D reconstruction from multiple images to recognize faces. They try to devise a strategy to tackle the problem of recognizing faces from images exhibiting strong pose (rotation and occlusion) and without prior knowledge (uncalibrated cameras, images from different sources). They do so by framing the recognition in the context of 3D structure from motion with missing data problems. Recently, there has been a strong trend towards using 3D information to verify and recognize faces. However, most of the state of the artworks is developed over 3D sensors (3D range finders, stereo). Here, they propose to do recognition by measuring the likelihood that one or more images were generated by a given 3D shape.
- Format: PDF
- Size: 295.7 KB