Markov Network-based Unified Classifier for Face Identification


- Author: Wonjun Hwang12, Kyungshik Roh1, and Junmo Kim2
1: Samsung Advanced Institute of Technology, Korea
2: Korea Advanced Institute of Science and Technology, Korea

- published at IEEE International Conference on Computer Vision (ICCV), 2013

- Paper: [PDF]

- Abstract
We propose a novel unifying framework using a Markov network to learn the relationship between multiple classifiers in face recognition. We assume that we have several complementary classifiers and assign observation nodes to the features of a query image and hidden nodes to the features of gallery images. We connect each hidden node to its corresponding observation node and to the hidden nodes of other neighboring classifiers. For each observation-hidden node pair, we collect a set of gallery candidates that are most similar to the observation instance, and the relationship between the hidden nodes is captured in terms of the similarity matrix between the collected gallery images. Posterior probabilities in the hidden nodes are computed by the belief-propagation algorithm. The novelty of the proposed framework is the method that takes into account the classifier dependency using the results of each neighboring classifier. We present extensive results on two different evaluation protocols, known and unknown image variation tests, using three different databases, which shows that the proposed framework always leads to good accuracy in face recognition.

- Selected Experimental Results (Unknown Variation Test)
  • Training Set: FRGC Ver 2.0 
  • Test Set: XM2VTS and BANCA databases
Wonjun Hwang,
Oct 23, 2013, 5:23 PM