A model based approach for pose estimation and rotation invariant object matching


ÜNSALAN C.

PATTERN RECOGNITION LETTERS, cilt.28, sa.1, ss.49-57, 2007 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 28 Konu: 1
  • Basım Tarihi: 2007
  • Doi Numarası: 10.1016/j.patrec.2006.06.003
  • Dergi Adı: PATTERN RECOGNITION LETTERS
  • Sayfa Sayıları: ss.49-57

Özet

Pose estimation has been considered to be an important component in many pattern recognition and computer vision systems. In this paper, we introduce a pose estimation method based on implicit polynomials. We also introduce a set of rotation invariant measures for object matching. We test both methods under colored i!pise, missing points, and affine transformation. Extensive testings indicate that our methods work fairly well under various disturbances. We also compare our methods with existing ones in the literature. (c) 2006 Elsevier B.V. All rights reserved.