Apparatus and method for generating image including multiple people
AssigneeSamsung Electronics Co., Ltd.
InventorsChang-Han Kim, Tae-Hwa Hong, Chung-Hoon Kim, Sung-Dae Cho
Life of this patent
Estimating the term…
Abstract
Provided is an apparatus and a method for generating an image including multiple people, in which images are photographed, face images are detected from each of the photographed images, a face score for each of the face images detected from the photographed images is calculated, it is determined whether each of the face scores calculated for each of the face images is at least equal to a predetermined threshold, and each of face images having a highest face score for each of the face images is synthesized, to output the synthesized image as a final image, in order to generate a most satisfactory photographed image for a user.
Claims23
An apparatus for generating an image including multiple photographed people, comprising:
- an image photographing unit for photographing images;
- a face image detection unit for detecting face images for each of the photographed images;
- a face score calculation unit for calculating a face score for each of the face images detected from each of the photographed images;
- an image determination unit for determining whether each of the face scores calculated for each of the face images is at least equal to a predetermined threshold and setting an image having a greatest number of face images having the face scores at least equal to the threshold among the photographed images as a reference image;
- and
- an image synthesis unit for synthesizing each of face images from other photographed images with the reference image, and outputting the synthesized image as a final image.
A method for generating an image including multiple photographed people, the method comprising the steps of:
- photographing images;
- detecting face images from each of the photographed images;
- calculating a face score for each of the face images detected from each of the photographed images;
- determining whether the face score detected for each of the face images is at least equal to a predetermined threshold;
- setting an image including a greatest number of images having the face scores at least equal to the threshold among the photographed images as a reference image;
- synthesizing each of face images from other photographed images with the reference image;
- and
- outputting the synthesized image as a final image.
Dependencies are inferred from the claim text.