Koninklijke Philips Corporation N.v.

Kees Van Zon

Nicolas Wadih Chbat

Haibo Wang

Konstantin Y Gagarin

Xiang Xiang

Disclosed techniques relate to identifying subjects in digital images. In various embodiments, digital image(s) that depict a subject in an area may be acquired. Portion(s) of the digital image(s) that depict a face of the subject may be detected as detected face image(s). Features of each of the detected face image(s) may be compared with features of each of a set of subject reference templates associated with a given subject in a subject reference database. Based on the comparing, a subject reference template may be selected from the set of subject reference templates associated with the given subject. Similarity measure(s) may then be determined between a given detected face image of the detected face image(s) and the selected subject reference template. An identity of the subject may be determined based on the similarity measure(s).

Subject identification systems and methods

A Dynamic Bayesian Network provides models that provides emulation of patient data.

Method and apparatus for deriving probabilistic models from deterministic ones

A method and apparatus for downscaling video images to a lower resolution (e.g. from HDTV to SDTV) is presented. The method comprising the steps of frequency domain anti-aliasing filtering and downscaling the first video signal in a first direction corresponding with a line direction in the first video signal to obtain a downscaled video signal, and spatial domain downscaling the downscaled video signal in a second direction perpendicular to the first direction to obtain the second video signal. The method and apparatus of the invention are suitable for efficient and high quality decoding both progressive and any-type encoded interlaced signals.

Growing community of inventors

Cold Spring, NY, United States of America

Kees Van Zon