JP2744394B2 - Realism image display device and realism image input / output device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image display apparatus and a face-to-face image communication requiring a high sense of reality, for example, a high-performance T
The present invention relates to a real-life image display device and a real-life image input / output device used for imaging and display devices such as a V-phone and a video conference.

[0002]

2. Description of the Related Art In recent years, with the development of image communication technology, TV telephones and video conference services in which remote locations are connected by an image and voice communication network have become widespread. In addition, among services using image communication, there is a demand for a terminal capable of obtaining a high sense of reality as a medical consultant, a legal consultant, or a communication terminal for VIP. Stereoscopic image display is being considered as a means of realizing high-presence image communication, but in order to obtain natural stereoscopic display, it is necessary to simultaneously transmit several tens of image information, and a large amount of communication capacity is required. Desired. Furthermore, when observing stereoscopic images for a long time,
It is necessary to overcome problems such as problems such as eye fatigue.

As a method for displaying an image having a sense of depth, which is a relatively simple structure, and which is a cause of a high sense of reality, a display method in which a person appears to be raised above the background has been devised. As a method that is currently in practical use, there is a magic vision method in which a half-mirror is used to combine and display a virtual image of a person in front of an actual background set.

FIG. 3 shows the configuration of the prior art. The observer 31 observes the real image 33 projected by the projector 32 as a virtual image person image 35 which is a reflection image of the half mirror 34. Since the virtual image person image 35 is observed at a position overlapping the real background 36, the person is observed as if it were floating in the space in the background 36.

[0005]

When the Magic Vision system is used for communication, the background 36 is a set of real objects,
The fixed situation limits the situations that can be used. Furthermore, since the human image is displayed as a virtual image, the overlapping portion between the human image and the background can be seen, and the illumination of the background 36 must be darkened to hide this. And there is a problem that background information is not utilized.

[0006]

In order to solve this problem, according to the present invention, a virtual image human image cut out from an arbitrary background and a background image obtained by combining the human image portion as a shadow image with an arbitrary background have different depths. Display on two surfaces. At this time, the virtual image person image is displayed in front of the background image so as to overlap the person shadow portion of the background image. Further, the human shadow portion in the background image is scrolled according to the left-right movement position of the observer's head.

In an image input / output device used as a face-to-face image communication terminal, a camera for photographing a person is installed behind a virtual image of a person.

[0008]

According to the present invention, an arbitrary background image corresponding to a communication situation or the like is displayed as a background image, and a portion observed as an overlap between a human image and a background is extracted from the background image by image processing to an observer, and is output as a black signal. Since the overlap of the images has been removed, both the person and the background images can be displayed brightly. In addition, by scrolling the shadow of the overlapped portion to the left or right according to the observer's observation position, the presence of motion due to the reproduction of motion parallax can be improved. Improvements are also possible. In addition, since a camera for photographing can be arranged between the person image and the background image, it is possible for the interlocutors to communicate with each other while seeing a natural person image whose eyes match. Also, unlike the stereoscopic display method using binocular parallax, the display surface of the human image and the display surface of the background are separated, so that a natural sense of depth can be felt, and problems such as eye fatigue due to long-term observation also occur. Absent.

[0009]

FIG. 1 shows a configuration example of the present invention. The virtual image generation and display unit 10 is a unit that generates and displays a virtual image human image 16 from a horizontally inverted image of the transmitted human image using a half mirror or the like. The composite background image generation means 11 generates a composite background image 17 in which a portion where a human image is present is replaced with a black image from the background image. Synthetic background image display means 12
Displays the composite background image 17 behind the virtual image person image 16 as viewed from the observer 18. The TV camera 13
The observer 18 is imaged. The movement detecting means 14 is a means for detecting the position of the head of the observer 18 in the image input from the television camera 13 and detecting the movement of the position.
The scroll control unit 15 converts the shadow image 19 in the composite background image 17 generated by the composite background image generation unit 11 into a virtual image human image according to the movement position of the observer's 18 head detected by the movement detection unit 14. The portion of the shadow image 19 is scrolled so as to be located at a position overlapping with the 16 person portions.

[0010] The television camera 13 is installed behind the virtual image person image 16 and an input image from this is transmitted to the communication partner device. The movement of the head of the observer 18 can be detected by an input from the television camera 13, but may be detected by other means using, for example, infrared rays.

FIG. 2 is a block diagram of an embodiment of the present invention. From the image transmitted from the opposite terminal, a person image is cut out by the person image extracting unit 20. The extraction of the person image by the person image extraction unit 20 can be performed by, for example, a method described in the following document. [References] Sakai, Takita, Nakamura: "A Method for Extracting Moving Regions", Technical Report of the Institute of Television Engineers of Japan, vol. 16, N
O. 32, pp19-23.

The left-right reversing unit 21 reverses the left and right of the cut-out person image and sends the image to a person projector 22. The projector 22 projects the person image on the screen 23. This image is observed by the observer 18 after being reflected by the half mirror 24. Therefore, the observer 18 observes the virtual image person image 16.

On the other hand, an arbitrary background image is input to the shadow image synthesizing scroll unit 25. The input background image may be an image transmitted from the opposite terminal or another previously prepared image. The background image is synthesized by the shadow image synthesizing scroll unit 25 with the shadow image in which the person region of the virtual image person image 16 is set as a black signal, and is sent to the background projector 26. The projector 26 projects the image so as to form an image on a screen 29 disposed behind the virtual image person image 16.

Therefore, the observer 18 moves the virtual image person image 16 forward and the synthesized background image 1 with the person portion missing behind.
7 will be observed at the same time. The image of the observer 18 captured by the television camera 13 is input to the image input unit 27 and transmitted to the opposite terminal. On the other hand, from this image,
The head position detection unit 2 determines whether the observer 18 has moved left and right.
8 to detect. The detection of the head position can be performed, for example, by using a method described in the following document. [References] Nagashima, Nakatani, Hisaki: "Study of Human Image Detection and Tracking Method in Video Signal", IEICE Research Report, IE87-87, pp17-24 (1987).

When the movement of the head position is detected by the head position detecting section 28, the coordinates of the head position are notified to the shadow image synthesizing scroll section 25. Shadow image composition scroll unit 2
Reference numeral 5 denotes a case where the shadow image of the person to be combined with the background image is scrolled left and right in accordance with the head position coordinates, and the shadow image of the combined background image 17 becomes the person image of the virtual image person image 16 when viewed by the observer 18. Are synthesized so as to overlap with.

The embodiment shown in FIG. 2 is an example of an image input / output device that can be used for a TV phone or the like, but an image display device that performs only display can be similarly constructed.

[0017]

As described above, according to the present invention,
By reproducing the motion parallax, the sense of reality can be improved, and both the person and the background image can be displayed brightly. Since the display surface of the human image and the display surface of the background are separated, a natural sense of depth can be felt, and there is no problem such as eye fatigue due to long-term observation. Further, since a camera for photographing can be arranged between the person image and the background image, it is possible to communicate while viewing a natural person image in face-to-face image communication.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration example of the present invention.

FIG. 2 is a configuration diagram of an embodiment of the present invention.

FIG. 3 is a diagram illustrating an example of a magic vision system.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Virtual image generation and display means 11 Synthetic background image generating means 12 Synthetic background image display means 13 Television camera 14 Movement detecting means 15 Scroll control means 16 Virtual image person image 17 Synthetic background image 18 Observer 19 Shadow image

 ──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-62-223720 (JP, A) JP-A-5-46161 (JP, A) JP-A-5-241540 (JP, A) 74292 (JP, U)

Claims (2)

(57) [Claims] 1. An image display device for displaying a human image and a background image in a superimposed manner, a virtual image generating and displaying means for generating and displaying a virtual image human image from a left-right inverted image of the human image, and a human image from the background image. A composite background image generating means for generating a composite background image in which a portion where is present is replaced with a black image, and the composite background image is formed so that the human image and the portion of the composite background image in which the black image is replaced overlap. Means for displaying behind the virtual image human image, means for detecting the position of the observer's head, and a human shadow part replaced with a black image in the composite background image according to the moving position of the observer's head. And a means for scrolling. 2. An image input / output device used for face-to-face image communication, wherein a virtual image generation / display means for generating / displaying a virtual image person image from a horizontally inverted image of a transmitted person image, and said person image from a background image A composite background image generating means for generating a composite background image in which a portion where is present is replaced with a black image, and the composite background image is formed so that the human image and the portion of the composite background image in which the black image is replaced overlap. Means for displaying behind the virtual image human image, means for detecting the position of the observer's head, and a human shadow part replaced with a black image in the composite background image according to the moving position of the observer's head. A realistic image input / output device comprising: means for scrolling; and means for photographing a person image to be transmitted to a partner device by a camera installed behind the virtual image person image.