TY - CHAP A1 - Gräßl, Christoph A1 - Deinzer, Frank A1 - Niemann, Heinrich ED - Krasnoproshin, V. ED - Ablameyko, S. ED - Soldek, J. T1 - Continuous Parametrization of Normal Distributions for Improving the Discrete Statistical Eigenspace Approach for Object Recognition T2 - Pattern Recognition and Information Processing Y1 - 2003 UR - https://d1wqtxts1xzle7.cloudfront.net/42129001/Continuous_parametrization_of_normal_dis20160205-16346-rayto3-libre.pdf?1454667530=&response-content-disposition=inline%3B+filename%3DContinuous_parametrization_of_normal_dis.pdf&Expires=1696432558&Signature=TmoSjs27NuupbCgHp6Y6aZdGGychAuKRf66hsbjelx-Zxe3FPaK25CiMC73SL~I4uz0DbjQN3ZwmFYwlgMuJRRcgO6AIaac4vNtJHfly-lOPIhXYqdsA3w3VwDeFU2SzEWZ7KEdn~rHjk4ZK8TZ6JOGzbO4CIe5SnOyhwUJL7w9o6~UwGGvtHKtNYOugvbwMdDr7iOanVhRQvMT5Fr1AEp5-mDS9VAg~7fkMD39pc-9vRBj0-4eS4tBQQFApJeYR244n7e~4wRClNqZp2lfn5kPjf-df5xj6iXBIUe4EKt53cR~ZFvE96N405yH14wGiTpGd1Vtx5VC-J0mS4o9XeQ__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA VL - 1 SP - 73 EP - 77 ER - TY - CHAP A1 - Gräßl, Christoph A1 - Deinzer, Frank A1 - Mattern, F. A1 - Niemann, Heinrich T1 - Improving Statistical Object Recognition Approaches by a Parameterization of Normal Distributions T2 - 6th German-Russian IAPR Workshop on Pattern Recognition and Image Understanding Y1 - 2003 SP - 38 EP - 41 ER - TY - CHAP A1 - Gräßl, Christoph A1 - Deinzer, Frank A1 - Mattern, F. A1 - Niemann, Heinrich T1 - Improving Statistical Object Recognition Approaches by a Parameterization of Normal Distributions T2 - 6th German-Russian IAPR Workshop on Pattern Recognition and Image Understanding N2 - As statistical approaches play an important role in object recognition, we present a novel approach which is based on object mod- els consisting of normal distributions for each training image. We show how to parameterize the mean vector and covariance matrix independently from the interpolation technique and formulate the classification and localization as a continuous optimization problem. This enables the computation of object poses which have never been seen during training. For interpolation, we present four different techniques which are compared in an experiment with real images. The results show the benefits of our method both in classification rate and pose estimation accuracy. Y1 - 2004 UR - https://www.researchgate.net/profile/Frank-Deinzer/publication/215748368_Improving_Statistical_Object_Recognition_Approaches_by_a_Parameterization_of_Normal_Distributions/links/0912f51017f34be4cb000000/Improving-Statistical-Object-Recognition-Approaches-by-a-Parameterization-of-Normal-Distributions.pdf SN - 1054-6618 VL - 14 IS - 2 SP - 222 EP - 230 ER -