KR20100060176A - Apparatus and method for compositing image using a face recognition of broadcasting program - Google Patents

Abstract

PURPOSE: An apparatus and a method for compositing an image using the face recognition of a broadcasting program are provided to automatically synthesize an image of a viewer in recognized face after an image of the broadcasting program or face recognition of a still image, thereby inducing interest of the viewer. CONSTITUTION: A face recognition unit(112) recognizes a face of individual in a frame of video data. The face recognition unit determines an area of recognized first face. The face recognition unit matches a recognized second face area in the recognized first face area after face recognition of inputted face image. A control unit(106) synthesizes the recognized face in the face image after removal of an image of the recognized first face area.

Description

Apparatus and method for image synthesis through face recognition of broadcasting program {APPARATUS AND METHOD FOR COMPOSITING IMAGE USING A FACE RECOGNITION OF BROADCASTING PROGRAM}

The present invention relates to an image correction technology using an image receiving and output apparatus. In particular, after recognizing a face image of a specific actor in a recorded image or a still image recorded through the image receiving and output apparatus, the viewer's face is recognized on the recognized actor's face. An apparatus and method for synthesizing an image through face recognition of a broadcast program suitable for automatically synthesizing an image are provided.

As is well known, the image receiving and outputting apparatus receives a broadcasting signal transmitted from each broadcasting station through an antenna or a cable, and displays a broadcasting channel desired by a user through tuning and decoding on the received signal.

In addition to the basic broadcast reception and output, such a video receiving and output device has a scheduled recording function of a received broadcast or a time machine function that enables recording and replaying of a broadcast a while ago, an IPTV and two-way broadcasting service function through an Internet connection, and Various additional functions such as playing MP3 files and playing video files are also included.

As such, the image receiving and output device not only provides a conventional broadcast program but also provides various additional functions, thereby providing a user with pleasure based on various functions besides watching a broadcast.

In the conventional image receiving and output apparatus operating as described above, the received broadcast program can be displayed or recorded and stored, but since the self-editing of the broadcast program cannot be performed, the actors appearing in the broadcast program In order to synthesize one's face or another person's face on the face of the user, the recorded broadcast program data must be transferred to a computer and edited therein, and there is no special way to implement it in the image receiving and output device.

Accordingly, the present invention recognizes a face image of a specific actor in a recorded image or a still image recorded through an image receiving and output device, and then recognizes the face of a broadcast program that can automatically synthesize a viewer's face image on the recognized actor's face. Provided are an image synthesizing apparatus and method.

In addition, the present invention recognizes the face image of a particular actor in the recorded image or still image recorded through the image receiving and output device, and compares the matched face image of the viewer photographed from the camera module to the viewer's face An apparatus and method for synthesizing an image through face recognition of a broadcast program capable of compositing are provided.

According to an embodiment of the present invention, a device recognizes a face of a person in a frame of image data, determines a region of a face recognized in the frame, recognizes a face in an input face image, and is recognized in the face image. And a face recognition unit for matching the face region to the face region recognized in the frame, and a controller for removing the image of the face region recognized in the frame, and then overlaying and synthesizing the face recognized in the face image.

According to an embodiment of the present invention, a method of determining a face region by recognizing a face of a person in a frame of image data, recognizing a face in an input face image, and then recognizing a face region in the image in the frame Matching the recognized face region, and when matching between the face regions, removing the image of the face region recognized in the frame, and then overlapping and synthesizing the recognized faces in the face image.

In the present invention, the effects obtained by the representative ones of the disclosed inventions will be briefly described as follows.

According to the present invention, after performing a face recognition of a video or a still image of a broadcast program, by automatically synthesizing the viewer's image on the recognized face, the viewer can be an actor of his favorite pro, thus causing interest to the viewer. It can be effected.

Hereinafter, the operating principle of the present invention will be described in detail with reference to the accompanying drawings. In the following description of the present invention, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted. The following terms are defined in consideration of the functions of the present invention, and may be changed according to the intentions or customs of the user, the operator, and the like. Therefore, the definition should be made based on the contents throughout the specification.

According to the present invention, after recognizing a face image of a specific actor in a recorded image or a still image recorded through an image receiving and output device, the face image of the viewer is automatically synthesized. By using an image, or using an image captured by the camera of the image receiving and output device.

1 is a block diagram illustrating a structure of a viewer face synthesizing apparatus through face recognition according to an embodiment of the present invention.

Referring to FIG. 1, a viewer face synthesizing apparatus through face recognition includes a broadcast receiver 102, an image processor 104, a controller 106, a memory 108, a control signal input unit 110, and a face recognition unit 112. It includes, and the camera module 114 may be further included.

The broadcast receiver 102 receives a broadcast signal transmitted from each broadcast station through an antenna or a cable, and performs tuning on the received signal. That is, a broadcast signal of a radio frequency (RF) channel is received, and the demodulated broadcast signal is demodulated, and the demodulated transport stream TS is transmitted to the image processor 104.

In the image processor 104, a demultiplexer for demultiplexing a demodulated TS into video / audio / additional / channel information through the broadcast receiver 102, and parsing for analyzing program information demultiplexed in the demultiplexer. And a decoding unit for decoding the demultiplexed video information. After that, the decoded signal is displayed on the screen.

The controller 106 controls the respective functional blocks, and transmits the decoded broadcast data from the image processor 104 to the memory 108 to be stored. In addition, the control signal input through the control signal input unit 110 receives the control function corresponding to the control signal.

The memory 108 stores broadcast data and viewer image data transmitted from the controller 106, and stores a face recognition program and an image composition program for face recognition, and stores the program in the face recognition unit 112 under the control of the controller 106. Let this be done.

The control signal input unit 110 receives control signals (eg, menu operation, channel change, volume control, etc.) for various operations for controlling a device or menu interface control transmitted from an operation panel or a remote controller of an image receiving and output device. It receives an input and delivers it to the controller 106.

The face recognition unit 112 recognizes a face from a broadcast image or a photographic image from at least one of the image processor 104, the controller 106, and the camera module 114, and cuts or recognizes a face of the recognized face. In the case of an image, face recognition and image synthesis are performed through face recognition for each frame.

In detail, the face recognition unit 112 analyzes the transferred image or image to recognize the face image, and checks information such as an angle and an overall area of the recognized face. Then, when receiving other image or image information to be synthesized to the recognized face, by analyzing the recognition of the face image, and confirms the information, such as the angle, the total area of the recognized face. In response to the face information to be synthesized with the first recognized face information, face information to be synthesized is reduced and enlarged to fit the first recognized face area, and the face to be synthesized within a preset error range is matched to the first recognized face area. If so, the face to be synthesized is inserted into the first recognized face area.

However, if the face to be synthesized does not match within the preset error range, the face recognition unit 112 transmits that the face is not matched to the control unit 106, and the control unit 106 requests input of other face data on the screen. Or display a composite fire message.

That is, through this, it is possible to synthesize the face of a specific actor in the broadcast program into face data input by the viewer.

The camera module 114 is for capturing a viewer's face image or image information. First, the camera module 114 deletes the face area recognized by the broadcast program while showing the viewer's face to be captured on the screen through live view, and then horizontally flips it. By performing the above, the viewer's face is overlaid on the live view screen.

That is, since only the face area recognized by the broadcast program is not in a state such as photographing a sticker, the viewer moves his or her face to and from the face area, and then takes a picture, and controls the photographed data. ) Or to the face recognition unit 112.

Of course, in general, it is also possible to deliver the photographed data through photographing and image capturing of the viewer.

This will be described in detail with reference to FIG. 3.

2 is a flowchart illustrating a synthesis process of a viewer's face through face recognition according to an embodiment of the present invention.

Referring to FIG. 2, in operation 200, broadcast data received through the broadcast receiving unit 102 and processed by the image processing unit 104 may be received by the controller 106 and stored in the memory 108 through operation 202.

In operation 204, the controller 106 transmits the image or image data of the specific part of the broadcast program stored from the memory 108 to the face recognition unit 112 under the control of the control signal. In 112, the face of a specific actor is recognized from the transferred image or image data, and the recognized face area is identified.

In step 206, the face region is recognized through the face image of the viewer input to the face recognition unit 112. The conversion is performed so that the viewer's face area matches the face of the specific actor.

Thereafter, in step 210, the synthesis is performed such that the viewer's face overlaps the face of the specific actor. In step 212, when the synthesis is completed, the process proceeds to step 214 to display the synthesized result on the screen.

However, when the synthesis is not completed in step 212, when the viewer does not like the synthesized state, the cancellation control signal is transmitted to the control unit 106, or the face recognition unit 112 itself transforms the viewer's face. If the viewer's face does not match the face of the specific actor within the error range, the controller proceeds to step 216 and requests the viewer another image through the control unit 106, and returns to step 206.

In step 206, when another face image is transferred to the face recognition unit 112, matching is performed on the face region recognized in the face region of the specific actor through face recognition.

3 is a flowchart illustrating a viewer face photographing procedure in an image receiving and output device according to an exemplary embodiment of the present invention.

Referring to FIG. 3, in operation 300, the camera module 114 analyzes a face recognized image through the face recognition unit 112 under the control of the controller 106, and recognizes the face of a specific actor in step 302. A horizontal flip on the drawn image. In other words, when the viewer shoots through the camera based on what the viewer sees on the screen, the left and right sides are reversed, so that the camera module 114 performs a horizontal flip so that the viewer can easily move the face to the face area of the specific actor. Implement so that

Accordingly, after removing the face region from the image of the specific actor horizontally flipped in step 304, the viewer's face, which is input through the live view of the screen, is shown to appear in the face region of the specific actor. In other words, other regions except for the face of a specific actor are displayed on the live view.

Accordingly, when the face region of the viewer is located within the preset error range in step 306, the user may automatically take a picture in step 308, and take a picture based on the control signal input by the user through the remote controller and the operation panel. . Accordingly, the viewer's face is synthesized in the face region of the specific actor which overlaps with the photographing. The synthesized result is displayed to the viewer on the screen in step 310.

However, if the matching result is not within the preset error range in step 306, the screen brightness is adjusted in step 312. If the error is not yet within the error range, if it is within the error range in the dark brightness state, the brightness may be adjusted. In addition, the overall saturation, etc. can be adjusted from dark to bright. You can also use voice guidance to provide announcements such as "Please move your location more", "Accurate. Take a picture, don't move."

If it is matched again at step 306, the process proceeds to step 308 to perform image composition.

As described above, according to the present invention, after recognizing a face image of a specific actor in a recorded image or a still image through an image receiving and output device, the present invention automatically synthesizes the viewer's face image on the recognized actor's face. The face image uses an input image or an image captured by a camera of an image receiving and output device.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiments, but is capable of various modifications within the scope of the invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by those equivalent to the scope of the claims.

1 is a block diagram showing the structure of a viewer face synthesizing apparatus through face recognition according to an embodiment of the present invention;

2 is a flowchart illustrating a synthesis process of a viewer's face through face recognition according to an embodiment of the present invention;

3 is a flowchart illustrating a viewer face photographing procedure in an image receiving and output device according to a preferred embodiment of the present invention.

<Explanation of Signs of Major Parts of Drawings>

102: broadcast receiving unit 104: image processing unit

106: control unit 108: memory

110: control signal input unit 112: face recognition unit

114: camera module

Claims (6)

Recognizing a face of a person in a frame of image data, determining a region of a first face recognized in the frame, recognizing a face in an input face image, and then recognizing the second face region recognized in the face image. Face recognition unit to match 1 face area, A control unit which removes the image of the first face area recognized in the frame and then synthesizes the overlapped face in the face image Image synthesizing apparatus through face recognition of a broadcast program comprising a. The method of claim 1, The face recognition unit, Matching to the first face region recognized in the frame by reducing and enlarging the second face region recognized in the face image, and when matching is completed within a preset error range, it is transmitted to the controller. Image synthesizing apparatus through face recognition of a broadcast program. The method of claim 1, The image synthesizing apparatus, After performing the horizontal flip of the image recognized in the frame transmitted from the face recognition unit, after removing the recognized first face area from the horizontal flipped image, the person to be photographed is displayed, and the person to be photographed is within a preset error range. When the camera is located in the recognized first face area, the image synthesizing apparatus through face recognition of a broadcast program, characterized in that it comprises a camera module for performing a shooting. Determining a first face area by recognizing a face of a person in a frame of image data; Recognizing a face in the input face image, matching the second face region recognized in the image with the first face region recognized in the frame; When matching between the first and second face regions, removing the image of the first face region recognized in the frame, and then overlapping the recognized face in the face image to synthesize Image synthesis method through the face recognition of a broadcast program comprising a. The method of claim 4, wherein The matching process is, Performing matching to the first facial region recognized in the frame by reducing and enlarging the second facial region recognized in the face image; When matching is possible within a preset error range, a process of performing a synthesis Image synthesis method through the face recognition of a broadcast program comprising a. The method of claim 4, wherein The input face image is, Performing a horizontal flip of the image recognized in the frame by the camera module; Removing the recognized first face area from the flipped image and showing a person to be photographed; Performing photographing when the person to be photographed is located in the recognized first face area within a preset error range Image synthesis method through the face recognition of a broadcast program, characterized in that the image generated through.

Images