摘要:
A face detection device comprises: a judgment area determining operable to judgment areas in an inputted image; and a face judgment unit operable to judge existence of a face image greater than a first predetermined size in the judgment areas. According to face directions, the judgment area determining unit determines a ratio of the inputted image to the judgment areas. The characteristics of face images can be effectively utilized.
摘要:
A stream reproducing device performing automatic viewing of a non-viewed period of a stream. The stream reproducing device comprises a camera a person detecting unit detecting the viewer based on the output of the camera, a viewing information generating unit generating viewing information based on the person detecting unit, and a reproduction control unit receiving supply of a stream, to control reproduction of the stream. The person detecting unit detects each of the viewers by classifying based on the output of the camera. The viewing information generating unit generates the viewing information for each of the viewers based on the detection result of each of the viewers. The viewing information generated by the viewing information generating unit is related to a time stamp of the stream to identify a non-viewed period, for which each viewer does not view the reproduced result of the stream.
摘要:
A face detection device comprises: a judgment area determining operable to judgment areas in an inputted image; and a face judgment unit operable to judge existence of a face image greater than a first predetermined size in the judgment areas. According to face directions, the judgment area determining unit determines a ratio of the inputted image to the judgment areas. The characteristics of face images can be effectively utilized.
摘要:
A viewing intention judging device comprises an input image control unit, an operation input unit, an element-image generating/selecting unit operable to generate a user interface element image and to handle selection, a composing unit operable to compose the user interface element image and an inputted image, a display unit operable to display the composed image, a person state acquiring unit operable to acquire information of a person near the display unit, a viewing intention judging unit operable to judge a viewing intention of the person, and an operation control unit operable to control the element-image generating/selecting unit and the operation input unit when a judgment result indicates transit from a state with viewing intention to a state without viewing intention or vice versa.
摘要:
A tracking processing unit (137) performs a search in a frame using a color feature amount of an object and determines a target area estimated to include the image of the object based on the first evaluation value indicating a degree of matching between a search result and the object. An update detection circuit (139) performs a search in the frame using a shape feature amount of the object, calculates a second evaluation value, and determines an update area for updating a first feature amount based on the second evaluation value indicating a degree of matching between the search result and the object. An update determination circuit (140) determines whether or not to update the first feature amount. A tracking result drawing circuit (144) draws an outer frame of the target area when determined not to update, and draws that of the update area when determined to update.
摘要:
A tracking processing unit (137) performs a search in a frame using a color feature amount of an object and determines a target area estimated to include the image of the object based on the first evaluation value indicating a degree of matching between a search result and the object. An update detection circuit (139) performs a search in the frame using a shape feature amount of the object, calculates a second evaluation value, and determines an update area for updating a first feature amount based on the second evaluation value indicating a degree of matching between the search result and the object. An update determination circuit (140) determines whether or not to update the first feature amount. A tracking result drawing circuit (144) draws an outer frame of the target area when determined not to update, and draws that of the update area when determined to update.
摘要:
A 3D image interpolation device performs frame interpolation on 3D video. The 3D image interpolation device includes: a range image interpolation unit that generates at least one interpolation range image to be interpolated between a first range image indicating a depth of a first image included in the 3D video and a second range image indicating a depth of a second image included in the 3D video; an image interpolation unit that generates at least one interpolation image to be interpolated between the first image and the second image; and an interpolation parallax image generation unit generates, based on interpolation image, at least one pair of interpolation parallax images having parallax according to a depth indicated by the interpolation range image.
摘要:
This image capture device is configured to generate a restored image by reducing a motion blur that has been induced by a camera shake during shooting. The device includes: an image sensor 10; an optical system 20; and an image processing section 220 which processes an image signal that has been obtained by the image sensor 10. The image processing section 220 includes: a blur kernel determining section which determines a blur kernel that defines the camera-shake-induced motion blur of the image captured; and an image restoration section which generates the restored image. The blur kernel determining section changes the size of the blur kernel by reference to information that affects the degree of the blur of the image such as the zoom power of the optical system 20.
摘要:
This 3D image capture device includes a light-transmitting section 2 with m transmitting areas (where m is an integer and m≧2) and a solid-state image sensor 1. The sensor 1 has unit elements, each of which includes n photosensitive cells (where n is an integer and n≧m) and n transmitting filters that face those photosensitive cells. If the wavelength is identified by λ, the transmittances of transmitting areas C1 and C2 are identified by Tc1(λ) and Tc2(λ), the transmittances of two transmitting filters are identified by Td1(λ) and Td2(λ), and the interval of integration is the entire visible radiation wavelength range, ∫Tc1(λ)Td1(λ)dλ>0, ∫Tc1(λ)Td2(λ)dλ>0, ∫Tc2(λ)Td1(λ)dλ>0, ∫Tc2(λ)Td2(λ)dλ>0, and ∫Tc1(λ)Td1(λ)dλ∫Tc2(λ)Td2(λ)dλ≠∫Tc2(λ)Td1(λ)dλ∫Tc1(λ)Td2(λ)dλ are satisfied.
摘要:
The 3D image capture device includes a light-transmitting section with n transmitting areas (where n is an integer and n≧2) that have different transmission wavelength ranges and each of which transmits a light ray falling within a first wavelength range, a solid-state image sensor that includes a photosensitive cell array having a number of unit blocks, and a signal processing section that processes the output signal of the image sensor. Each unit block includes n photosensitive cells including a first photosensitive cell that outputs a signal representing the quantity of the light ray falling within the first wavelength range. The signal processing section generates at least two image data with parallax by using a signal obtained by multiplying a signal supplied from the first photosensitive cell by a first coefficient, which is real number that is equal to or greater than zero but less than one.