The Quotient Image: Class-Based Re-Rendering and Recognition with Varying Illuminations

  • Authors:

  • Amnon Shashua;Tammy Riklin-Raviv

  • Affiliations:

  • The Hebrew Univ., Jerusalem, Israel;The Hebrew Univ., Jerusalem, Israel

  • Venue:
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Year:
  • 2001

Quantified Score

Hi-index 0.14

Visualization

Abstract

The paper addresses the problem of 驴class-based驴 image-based recognition and rendering with varying illumination. The rendering problem is defined as follows: Given a single input image of an object and a sample of images with varying illumination conditions of other objects of the same general class, re-render the input image to simulate new illumination conditions. The class-based recognition problem is similarly defined: Given a single image of an object in a database of images of other objects, some of them are multiply sampled under varying illumination, identify (match) any novel image of that object under varying illumination with the single image of that object in the database. We focus on Lambertian surface classes and, in particular, the class of human faces. The key result in our approach is based on a definition of an illumination invariant signature image which enables an analytic generation of the image space with varying illumination. We show that a small database of objects驴in our experiments as few as two objects驴is sufficient for generating the image space with varying illumination of any new object of the class from a single input image of that object. In many cases, the recognition results outperform by far conventional methods and the re-rendering is of remarkable quality considering the size of the database of example images and the mild preprocess required for making the algorithm work.