JP2005301693A - Video editing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a moving image editing system capable of performing a physical operation by a finger gesture.
On a moving image editing screen projected on a screen 150 from a projector 130 connected to an image projection computer 110, a user 160 performs each operation of moving image editing with a finger gesture. The camera 140 connected to the image processing computer 120 captures the gesture, and the image processing computer 120 recognizes the position of the finger. The image projection computer 110 recognizes the meaning of the gesture from the position of the finger of the user 160, performs image editing processing according to the gesture, and projects the moving image editing screen onto the screen 150 again. With the above interface, the problem of the conventional video editing system that was operated with a mouse etc. was solved.
[Selection] Figure 1

Description

  The present invention relates to a moving image editing system using gesture recognition.

In recent years, with the widespread use of digital video and digital cameras, it has become possible to save moving images and still images taken by the user on a PC for personal editing. When the photographic equipment was analog, video editing work was done using unique devices equipped with input devices with various dials and buttons (hereinafter referred to as “analog input devices”) such as jog shuttles and sliders. It was. Since these input devices are designed so that all the operation information is always visible, it seems to be complicated at first glance. Therefore, it is considered difficult for beginners to handle, but as the user continues to use it, the operation method is learned by a physical sense, and intuitive operation becomes possible. Therefore, it can be said that it has the feature that operation in real time becomes possible as the skill of the user increases.
Recently, software for editing video data on a PC, such as Non-Patent Document 1, Non-Patent Document 2, and Non-Patent Document 3, has been widely used. The reason for this is that the editing work can be performed without preparing the above-mentioned unique devices, and if the user is used to handling the mouse, it can be used immediately.
However, the following can be considered as problems in using the mouse on a GUI application such as the above-described conventional moving image editing software for operations that have been performed using analog input devices.

(1) Since the physical movement as realized by the analog input device cannot be performed, the amount of operation and the like cannot be physically understood. For this reason, physical experience is not reflected in operation, and skill is not reflected in operation.
(2) In analog input devices, it is possible to know in real time how the input is reflected in the output while performing simultaneous operation with a plurality of fingers, but generally pointing can be performed with an input using a mouse. The number of places is limited to one. For this reason, a plurality of operations cannot be performed simultaneously.
(3) In conventional GUI applications, a pull-down menu is often used, and it is difficult to always make a plurality of information necessary for operation visible.
(4) When using a large display as a monitor, the mouse is not always an appropriate input. That is, when a large display is used, it is necessary to repeat the relative movement of the mouse many times in order to move the mouse cursor from end to end of the screen. In order to solve this, it is possible to set the speed of the trajectory drawn by the mouse, but in that case, it is difficult to move the mouse finely.

As a method for solving these problems, it is conceivable to use a large display and construct an interface having the goodness of an analog input device as its input device.
The inventors have been researching EnhancedDesk, which is a desk-type real world oriented interface (for example, Non-Patent Document 4). EnhancedDesk can input finger positions, gestures, and object positions into the system, and as its application, it has realized a drawing tool that uses both hands and a system that supports meetings with a small number of people (for example, non-patent literature) 5, non-patent document 6). In these systems, physical operation is possible by using fingers as input means, and the system can be operated more intuitively while physically feeling the designation of the amount. In addition, since the positions of individual fingertips are recognized, it is possible to perform pointing at 10 locations at the same time, and it is possible to operate a plurality of digital information at the same time. Furthermore, since a desk is used as the projection surface of the projector, an environment as a large display is realized. From these points, it seems that EnhancedDesk is one of the effective environments in which the work that has conventionally been used for analog input devices is performed digitally using a large display.
However, a video editing system that solves the problems of conventional video editing software (for example, Non-Patent Documents 1 to 3) and can instruct video editing work with gestures has not been developed, and the above problems are solved. It has not been.

Adobe, Premiere. Http://www.adobe.com/products/premiere/ Apple, Final Cut Pro. Http://www.apple.com/finalcutpro/ Apple, iMovie. Http://www.apple.com/imovie/ Takanori Kobayashi, et al., “Real-time fingertip recognition interface using infrared image for EnhancedDesk”, Japan Software Science Society WISS 1999, pp.49-54, 1999. Chen Zhao, et al., “Drawing System with Two-handed Direct Operation in Desk Type Interface”, Japan Software Science Society WISS 2001, pp.179-184, 2001. Shinichiro Nagashima, et al., “EnhancedTable: Meeting Support System for Paper”, Japan Software Science Society WISS 2002, pp.111-116, 2002.

  An object of the present invention is to provide a video editing system that can be reflected in the operational intuition and skill workability, which is an advantage of an analog input device, and that can perform multiple simultaneous operations. It is to solve the problem of video editing software.

In order to solve the above-mentioned problems, the present invention displays a moving image editing screen, and displays the moving image editing screen in a moving image editing system that performs a moving image editing operation by inputting the user's gesture to the camera with a camera. A display unit; an image input unit that inputs an image captured by the camera; a finger recognition unit that recognizes a position of the user's finger from the input image; and a meaning of the user's gesture from the position of the user's finger. A moving image editing system comprising: a gesture recognizing unit; and a moving image editing unit that edits a moving image corresponding to the gesture, and displaying a moving image editing result as a moving image editing screen.
The video editing system further includes an initial screen storage unit that stores an image before the user's finger is captured, and the finger recognition unit compares the input image with the image from the initial screen storage unit. It is also possible to acquire the user's finger area, acquire the position of the user's fingertip by circular template matching, and recognize the position of the user's finger.
Further, the circular template matching in the finger recognition unit may be performed on a square region centered on the center position of the user's hand.
The video editing display means arranges a plurality of videos in a horizontal row on the video editing screen, and when each video is played back, the playback position is delayed by a certain time as the adjacent videos go from right to left. It can also be characterized in that it is played back in a live state.

When the position of the finger of the user is changed from a state where the hand is opened at the same place on the editing screen to a state where the hand is closed, the gesture recognition unit It can also be characterized by recognizing that a video has been grabbed.
The gesture recognizing unit selects an image or a moving image displayed at the position of the user's finger when the finger is placed at the same place on the editing screen. It can also be characterized as recognizing that
The gesture recognizing means displays the user at the position of each hand when the finger positions of both hands of the user are in the state where one finger is put out at the same place on the editing screen. It is also possible to recognize that the selected image or moving image and the image or moving image displayed between them are selected. Furthermore, after the user selects an image or a moving image, when both hands are brought close to each other, it is possible to recognize that the selected image or moving image is connected to one.
Further, the gesture recognizing means, when the position of the user's finger moves to any one of up, down, left and right with the hand open, its moving direction, moving speed and moving distance are the amount of moving image editing. It can also be characterized by recognizing.
A program for causing a computer system to realize the function of the moving image editing system described above is also the present invention.

  The image editing system of the present invention uses a finger recognition system for input to the screen. For this reason, the user can perform each operation of moving image editing with gestures. According to the present invention, it is possible to perform input based on physical movements that a conventional analog input device has and a plurality of simultaneous operations, and to solve the problems of conventional moving image editing software.

Hereinafter, embodiments of the moving image editing system of the present invention will be described in detail.
<1. System configuration>
First, the system configuration of the moving image editing system of this embodiment will be described with reference to the system configuration diagram shown in FIG.
As shown in FIG. 1, the moving image editing system according to the present embodiment connects an image processing computer 120 that performs finger recognition and an image projection computer 110 that performs moving image editing processing and image projection onto a screen 150. A projector 130 is connected to the projection computer 110, and a camera 140 is connected to the image processing computer 120. The projector 130 and the camera 140 are installed above the screen 150. Here, the image acquisition area of the camera 140 and the projection range by the projector 130 are approximately the same. In the present embodiment, a case where projection onto the screen 150 is performed using the projector 130 will be described as an example. However, the screen may be displayed on an ordinary large display or the like without using the projector 130.
The user 160 performs each operation of editing a moving image by a gesture using fingers on the screen projected from the projector 130 onto the screen 150. The image on the screen 150 captured by the camera 140 is transmitted to the image processing computer 120, and the image processing computer 120 recognizes the position of the finger. By transmitting the recognized finger position to the image projection computer 110, interaction between an object (such as an image or a moving image) projected on the screen by the projector 130 and the finger is realized. That is, the image projection computer 110 recognizes the gesture of the user 160 based on the position of the finger sent from the image processing computer 120, performs each video editing process corresponding to the gesture, Are projected again from the projector 130 onto the screen 150.

A “gesture recognition program” is installed in the image projection computer 110 to perform a gesture recognition process for recognizing the meaning of the gesture of the user 160 from the position of a finger sent from the image processing computer 120. In addition, a “moving image editing application” is installed, and a moving image editing process is performed by an operation of the user 160.
On the other hand, the image processing computer 120 is equipped with a “hand recognition program” and an “image processing library”. The hand recognition program uses the image processing library to detect the position of the finger of the user 160 (screen 150) captured by the camera 140. Finger recognition processing for recognizing the upper position) is performed. The inventors have conventionally used hardware such as an image processing board for image processing for recognizing a fingertip. An advantage of performing image processing with hardware is that advanced image processing can be performed even with a low-spec computer. However, one processing board is required for one computer in order to perform processing by hardware. When the number of computers that newly perform image processing is increased, image processing boards are required for the increased number of computers, leading to lack of expandability. In recent years, since the processing speed of computers has become much faster, it is no longer necessary to rely on hardware for image processing, and conversely, the disadvantages of using hardware have become conspicuous.
Therefore, in this embodiment, for example, an image processing library OpenCV (Intel Open Computer Vision Library) capable of performing image processing by software is installed in the image processing computer 120. As a result, a system with high convenience and expandability can be obtained. Furthermore, since this library is open source, it is considered possible to build a more flexible system.

The camera 140 is an IEEE1394 camera, for example. Infrared cameras and CCD cameras can also be used, but since these cameras have analog input / output terminals, it is necessary to convert image data taken with digital signals into analog signals, import them into a computer, and return them to digital signals again. . Also, special hardware is required to capture input from the camera. Therefore, in this embodiment, these problems are solved by using an IEEE1394 camera. IEEE1394 is a new interface standard suitable for computer environments. Since it is excellent in real-time data transfer and inexpensive, it is easy to introduce and is now widely used. By using an IEEE1394 camera equipped with this interface, a more general-purpose system is considered.
In addition, since the present embodiment is intended for processing in real time, it is necessary to increase the speed of finger recognition. Therefore, when recognizing the finger area from the image of the camera 140, it is necessary to make it easy to take the difference between the screen 150 and the finger. For this reason, a white table that is a single color is used for the screen 150.
In the system configuration diagram of FIG. 1, two computers, the image processing computer 120 and the image projection computer 110, are prepared. However, the present invention is not limited to this. For example, an image processing computer or an image projection computer is used. A plurality of units can be used as necessary, and both image processing and image projection may be performed by a single computer.

<2. Finger recognition>
Next, a flow of finger recognition processing in the moving image editing system of the present embodiment will be described. As described above, the finger recognition processing is performed by the camera 140 using the image processing library (for example, the above-described OpenCV in the present embodiment) installed in the finger recognition program installed in the image processing computer 120. This is processing for recognizing the position of the finger of the user 160 (position on the screen 150).
Before starting the finger recognition processing, the initial screen of the moving image editing application is projected from the projector 130 onto the screen 150.
When the finger recognition process is started, first, an image of the screen 150 on which the initial screen is projected is acquired from the camera 140 and stored as an initial image. Thereafter, the finger region is acquired based on the difference between the image captured by the camera 140 in real time and the initial image stored above. After that, when the screen of the video editing application transitions due to, for example, video editing performed by the user 160 gesture, the screen after the transition is stored as an initial image, and the finger area is acquired again by the above-described method.
By acquiring the finger area, it is possible to recognize on which object (image, moving image, etc.) on the moving image editing screen projected on the screen 150 the user's 160 hand is placed.

In the present embodiment, gestures such as a state where the hand is opened, a state where the hand is closed, and a state where one finger is taken out are used. For this reason, it is necessary to recognize the fingertip of the user 160 on the screen 150.
For fingertip recognition, template matching using a circular template is used based on the fact that the contour of the fingertip shape is close to a circle. The template matching has a problem that the processing result is reliable, but the calculation cost is high. Therefore, in this embodiment, in order to increase the processing speed, the center of the palm is obtained in advance, and template matching is performed only on a square area of 60 pixels × 60 pixels centered on that point. The value of 60 pixels is a value that is empirically set as a size that allows all palms to be included in an image captured from the camera, but may be another appropriate value. In addition, since a circle is used as a template, template matching processing that is usually sensitive to rotation can be performed without any problem.
In the finger recognition process, the position of the finger of the user 160 is recognized as described above.
In this embodiment, since the IEEE1394 camera and OpenCV are used, the finger recognition processing speed is improved as compared with the case where image processing is performed by hardware using a conventional infrared camera or CCD camera.

<3. Video editing system using finger recognition>
In the moving image editing system of the present embodiment, the user's fingers are used as an input device that replaces the analog input device. Since the system can be operated without using a medium such as a mouse, it is considered that the user can operate more intuitively. As a library used in the moving image editing system of the present embodiment, for example, JMF (Java® Media Framework) 2.1.1 and QuickTime for Java® that are good at moving image files are used.

(3-1. Recognition of gestures)
The user 160 performs a gesture with a finger on the screen of the moving image editing application projected on the screen 150 and performs each operation of the moving image editing system of the present embodiment. FIG. 2 is an example of a screen of the moving image editing application. As shown in FIG. 2, a number of objects (images, moving images, etc.) are displayed on the moving image editing screen in the present embodiment, and the user performs a gesture with a finger directly above the desired object. The screen configuration of the moving image editing application shown in FIG. 2 will be described in detail later.

The gesture recognition process is performed by the image projection computer 110. As described above, when the position of the user's finger is recognized by the finger recognition system of the image processing computer 120, the image projection computer 110 receives information on the position of the finger, and the gesture recognition process performs the user's gesture. Recognize meaning. In the present embodiment, four types of meanings shown in FIGS. 3A to 3D are recognized.
(A) On the object projected on the screen, from a state where the finger is opened (311) to a close (312) gesture: It is recognized that the object is grasped.
(B) Gesture with one finger on the object projected on the screen: It is recognized that the object has been selected. When this gesture is performed with both hands, it is recognized that the object selected with the finger of the left hand to the object selected with the finger of the right hand is selected as the operation target.
(C) Gesture to move up / down / left / right with hand open: It is recognized as a gesture to pass the amount of movement (movement direction, movement distance, movement speed, etc.) to the system.
(D) Perform the above selection (b) with both hands (in the figure, the object 343 is selected with the left hand 341, and the object 345 is selected with the right hand 342). It is recognized that a certain object (343, 344, 345) is grouped into one (347). Here, the illustrated object 346 is not selected.
Note that the gesture recognition program according to the present embodiment uses the gesture recognition program when a certain time (for example, 0.5 seconds) elapses with the hand open or closed, or with one finger out. Recognize as Also, the above (a) to (d) are examples, and it is possible to give meaning to other gestures as well.

(3-2. Editing video with gestures)
In constructing a video editing system that performs video editing using finger recognition, in this embodiment, the following five main functions are selected from the operations performed in a conventional general video editing system for PC. The video editing system was implemented by incorporating the characteristics of analog input devices.
(A) Select (B) Move (C) Fast forward / rewind movie file (D) Cut movie (E) Link movie That is, if the movie editing system of this embodiment is used, the user can By combining the four gestures of (d) to (d), the moving image editing functions shown in (A) to (E) can be used.
Note that the moving image editing functions (A) to (E) described above are merely examples, and other functions can be similarly implemented.

  Hereinafter, referring to FIGS. 4 to 8 showing the screen of the video editing application shown in FIG. 2 and examples of operations on the screen, “cut out a necessary part from the original video file, The above operations (A) to (E) will be described with reference to an example of an operation of “concatenating and creating a new moving image file”. Note that the video editing process (A) to (E) and the screen display process itself performed by the video editing application of the present embodiment are the same as those of the conventional video editing software.

FIG. 2 is an example of a screen configuration of the moving image editing application of the present embodiment. The screen 200 is a screen projected on the screen (screen 150 shown in FIG. 1). The user performs a finger gesture on the screen 200. In each area 210 to 280 of the screen 200, an object (an image, a moving image, etc.) for a user to perform an operation is displayed.
An area 220 in the upper right of the screen is an area (hereinafter referred to as “clipDock”) that displays a moving image file that is a source edited by the moving image editing system of the present embodiment. The original moving image file (for example, a mov file) may be stored in advance in a specific directory of the image projection computer 110, for example, and read automatically when the application is started and displayed on the clipDock 220. . The user selects a moving image file to be edited from clipDoc 220 projected on the screen. In an area 230 under the clip Dock 220, the playback time of the moving image file displayed in the clip Dock 220 is displayed.
An area 250 at the center of the screen is an area for playing back and displaying a moving image file (hereinafter referred to as “flowViewer”). Perform a return. In the upper area 240 of the flow Viewer 250 (hereinafter referred to as “timeline Viewer 240”), the playback position (timeline) of the video file being played back is displayed.

An area 210 on the left side of the screen is an area for displaying an image being operated by the user, and is hereinafter referred to as “in / outViewer”. For example, in the case of a cutting operation, the image of the starting point of cutting specified by the user is displayed on the in / outViewer 210 until the end point is specified, so that the user can easily understand. An area 260 at the bottom center of the screen is an area for displaying the cut video file as a thumbnail or the like, and is hereinafter referred to as “clipViewer”. Actually, for example, when a cut operation is performed, the cut portion is stored as a new moving image file in another directory (a directory prepared for clipViewer).
The area 270 below the clipViewer 260 is an area for displaying what the user designates as a target of processing (for example, concatenation) among the moving image files displayed on the clipViewer 260, and is hereinafter referred to as “compileviewer”. Actually, for example, a file is moved from a directory (clipviewer directory) that stores the movie file (cut movie file) displayed on the clipViewer 260 to another directory (compileviewer directory). To do.
In the area 280 at the bottom of the screen, the total playback time of the moving image file in the compieviewer 270 is displayed.

(1) Selection of moving image file First, a moving image file obtained by shooting or the like is selected. As described above, the captured moving image file is stored in the image projection computer 110 in advance, and the moving image file is displayed in the clip Dock 220 at the upper right of the screen 200.
When the user performs a gesture of standing still for a certain period of time (for example, 0.5 seconds) with a single finger in clipDock 220, the video file displayed under the finger is selected, and a flow viewer 250 described later is selected. Playback starts. At this time, in order to make it easy to understand which file is being reproduced, the display method of the image displayed in the clipDock 220 may be changed. For example, the selected image is displayed with an actual color, and other files are displayed with a reddish image.

(2) Reproduction of moving image file In the editing operation of moving image, there is a cutting operation of cutting out a necessary part from the captured moving image file. This operation is performed by designating a necessary start point and end point in the moving image file. At that time, the user identifies the scene by comparing with the preceding and following scenes. However, in conventional general video editing software, there is often a single screen for playing and displaying video files, and moving, rewinding, and fast-forwarding video files are repeated many times while moving the timeline with the mouse. The scene is specified by.
Therefore, in the present embodiment, a flow viewer 250 is provided that reproduces a moving image file in a state where a plurality of moving image files to be edited are arranged side by side (for example, seven in this embodiment) on the same screen.
FIG. 4 shows a state of reproduction by the flowViewer 250. The same video file is reproduced and displayed on the flowViewer 250 with seven images from 410 to 470. Here, in the present embodiment, as the adjacent images move from right to left, the reproduction position is reproduced with a delay of a certain time (for example, 100 msec). This makes it appear to the user as if the video file is flowing from right to left. Therefore, the user can see the flow of the video file at a time, and by comparing with the scenes before and after, it is possible to quickly select the scene that the user wants compared to the conventional timeline method. It becomes. In addition, on the upper part of the flowViewer 250, a timelineViewer 240 for displaying the current playback position of the moving image file on a timeline or the like for each area is provided.

(3) Control of playback speed When a user controls playback of a video file (fast forward / rewind, etc.), the playback position of the video is at a glance with an interface using a slide bar implemented in conventional video editing software. On the other hand, there is an advantage that the slide bar is very small compared to the screen ratio, so pointing is difficult, and it is difficult to find the position of the scene accurately using the mouse. In order to make up for the problem, it is possible to input numerical values with the keyboard, but determining the playback position of the movie file by numerical input is not intuitive and can be said to be a very unnatural interface. .
One solution is to make the amount of movement of the mouse correspond to the display speed and display size of the moving image (for example, http://www.yugop.com/). In this method, the object is moved in the corresponding direction by moving the mouse left and right. At that time, the moving speed of the object increases as the object moves away from the center of the screen. Since the movement distance of the mouse corresponds to the movement speed of the object, it can be said that the operation method is easy for the user to understand.
Also in this embodiment, the reproduction of the moving image file is controlled by the moving direction and moving distance of the hand. When the hand is slid with the hand open (the gesture shown in FIG. 3 (c)), the playback speed and display size of the video being played are changed according to the moving direction and moving distance.

In the present embodiment, two types of interfaces are constructed as a control (fast forward / rewind) for reproducing a moving image file. One is a method of determining by the position where the hand moves. In this method, the playback speed is changed using the coordinates of the hand, with the center of the screen projected on the screen as the origin. For example, fast forward right from the center of the screen, rewind left from the center, slide your hand to the end of the screen to fast forward and rewind the video file, and as your hand approaches the center of the screen Reduce the speed.
The other is a method of changing the playback speed according to the moving speed of the slid hand. This is because the playback speed of the moving image file changes depending on the speed at which the hand is slid. Specifically, when the hand is slid to the right, the playback speed is increased, and when the hand is slid to the left, the playback speed is decreased. Both the rate of increase and the rate of decrease are increased / decreased depending on the speed at which the hand is slid. Thereby, the user can change the position of the scene quickly and physically.
FIGS. 5A to 5C show gestures for increasing the playback speed when the playback speed is increased by sliding the hand to the right and the playback speed is decreased by sliding the hand to the left. As shown in FIG. 5A, first, the left hand 510 is opened on the flowViewer 250 of the screen projected on the screen. Next, as shown in (b) and (c), when the left hand 510 is slid rightward with the left hand 510 open, the reproduction speed of the moving image file reproduced by the flowViewer 250 can be increased.

Further, the display size of the moving image files arranged in the flowViewer 250 is changed by a gesture of sliding the open hand up and down. For example, the initial value of the number of images to be displayed at the same time is set to seven as shown in FIG. 4, and as the open hand is slid upward, the image display size is reduced and the number of images to be displayed at the same time is increased. Conversely, the smaller the slide, the larger the display size and the number of images to be displayed simultaneously. This is convenient when it is desired to increase the number of scene selection candidates or when it is desired to display a large scene.
In this way, while it is possible to perform playback, fast forward, and rewind by physical operation (gesture) while comparing with the next scene at the flowViewer 250, the scene that the user seeks rather than the conventional slide bar or numeric input It is easy to find the position of sensuously.

(4) Cutting a moving image file When the user performs a “grab” gesture (the gesture in FIG. 3A described above) on the flowViewer 250, the start point and the end point of the cutting can be determined in a scene immediately below the hand. When the cut start point is designated, the image of the scene is displayed in the in / outViewer 210 of the screen 200 shown in FIG. 2, and when the end point is designated, the original movie file is automatically moved from the start point to the end point. Cut out the video. The clipped moving image is displayed on clipViewer 260 at the bottom of the screen.
Gestures in which the user designates the start point of cutting are shown in FIGS. Here, an operation for designating the start point is performed with the right hand 620. First, as shown in (a), the scene of the starting point is displayed on the flowViewer 250, and the right hand 620 is placed in an open state, and then the hand is closed as shown in (b). "Do a gesture. Thus, the start point of the cut is specified, and the image of the start point is displayed in in / outViewer 210 as shown in (c). Next, specify the end point with the same “grab” gesture.

(5) Concatenation and playback of cut out moving image files Next, a plurality of cut out moving image files are connected to create one moving image file.
First, the user performs a selection gesture (gesture shown in FIG. 3B) on the image in the clipViewer 260, selects a moving image file to be connected, and moves it to the compieviewer 270. It is also possible to return the moving image file displayed on the compieviewer 270 to the clipViewer 260 by performing the reverse gesture.
Also, as shown in FIGS. 7A to 7C, when a selection gesture is performed with both hands, both the selected files and all the files in between can be moved. In this case, as shown in FIG. 7A, images are selected one by one with the clip viewer 260 using the right hand 720 and the left hand 710 (the gesture shown in FIG. 3B). Next, as shown in (b), it is moved to completeViewer 270 with the fingers of both hands out. Then, among the video files displayed on clipViewer 260, there are three files in total: two files selected with the fingers of both hands of the user and one file displayed between the two files. As shown in 7 (c), it is displayed on the compieviewer 270 and deleted from the clipviewer 260.
By repeatedly performing the above gesture, the user can select only a moving image file to be connected from the cut moving image files.
In the lower right of the screen, the total playback time 280 of the moving image file displayed on the compieviewer 270 is automatically displayed.

Next, the moving image file moved to the compieviewer 270 is connected. On the compieviewer 270, the fingers of both hands are taken out one by one, a moving image file that starts and ends is selected, and a gesture (gesture shown in FIG. 3D) for bringing these fingers close is performed. Then, the moving image files stored in the compieviewer 270 are connected to create one moving image file.
Here, a moving image created by the file displayed on the compieviewer 270 can be reproduced and confirmed before actually connecting. The gesture of this operation is shown in FIGS. First, as shown in (a), with the fingers of the left hand 810 and the right hand 820, select the video file that starts and ends the playback, respectively (the video file selected with both hands and the video file between them are selected. That ’s true). Next, both hands are moved to the center 830 of the screen with the gesture (b). Then, as shown in (c), the moving image file selected in (a) is reproduced and displayed in the center 830 of the screen in the order in which they are displayed on the compieviewer 270 (for example, in order from the left). In this way, it is possible to confirm in advance a preview of the resulting video.

If the moving image editing system of the present embodiment is used, the user can edit the moving image file with a physical sense as described above.
Note that the above is an example, and in the same manner, the video editing functions other than the above five (A) to (E) are realized by combining the four gestures (a) to (d). Can be made. Furthermore, if a gesture other than those shown in (a) to (d) is recognized, it is possible to realize more moving image editing functions.

1 is a system configuration diagram of a moving image editing system of the present embodiment. It is the figure which showed the structural example of the screen projected on a screen. It is the figure which showed the gesture recognized by this embodiment. It is the figure which showed an example of the display of flowViewer. It is the figure which showed the gesture which raises the reproduction speed of a moving image file. It is the figure which showed the gesture which designates the starting point of cutting of a moving image file. It is the figure which showed the gesture which selects and moves several moving image files at once. It is the figure which showed the gesture which puts together several moving image files into one.

Claims (10)

In the video editing system that displays the video editing screen and inputs the user's gesture for it with the camera to perform the video editing operation.
Video editing display means for displaying the video editing screen;
Image input means for inputting an image taken by the camera;
Finger recognition means for recognizing the position of the user's finger from the input image;
Gesture recognition means for recognizing the meaning of the user's gesture from the position of the user's finger;
A moving image editing means for performing moving image editing corresponding to the gesture, and displaying a result of moving image editing as a moving image editing screen. The video editing system according to claim 1,
Furthermore, it comprises an initial screen storage means for storing an image before the user's fingers are captured,
The finger recognition means compares the input image with the image from the initial screen storage means to obtain a user finger area, obtains the position of the user fingertip by circular template matching, and obtains the user finger A video editing system that recognizes the position of the video. The video editing system according to claim 2,
The moving image editing system, wherein the circular template matching in the finger recognition unit is performed on a square region centered on a center position of a user's hand. In the moving image editing system according to any one of claims 1 to 3,
The video editing display means arranges a plurality of videos in a horizontal row on the video editing screen, and when each video is played back, the playback position is delayed by a certain time as the adjacent videos go from right to left A video editing system characterized by being played back on. In the moving image editing system according to any one of claims 1 to 4,
When the position of the finger of the user is changed from a state where the hand is opened at the same place on the editing screen to a state where the hand is closed, the gesture recognition unit A video editing system characterized by recognizing that a video has been grabbed. In the moving image editing system according to any one of claims 1 to 5,
The gesture recognition means, when the position of the finger of the user is a state where one finger is put out at the same place on the editing screen, the user selects an image or a video displayed at the position A video editing system characterized by recognition. In the moving image editing system according to any one of claims 1 to 6,
The gesture recognition means is an image in which the user is displayed at the position of each hand when the positions of the fingers of both hands of the user are in the state where one finger is put out at the same place on the editing screen. Or a moving image editing system that recognizes that a moving image and an image or moving image displayed between them are selected. The moving image editing system according to claim 7,
The moving image editing system according to claim 1, wherein the gesture recognizing unit recognizes that the selected image or moving image is connected to one when the user brings both hands close after selecting the image or moving image. In the moving image editing system according to any one of claims 1 to 8,
The gesture recognizing means recognizes the moving direction, moving speed, and moving distance as the amount of video editing when the position of the user's finger moves up, down, left, or right with the hand open. A video editing system characterized by The program for making a computer system implement | achieve the function of the moving image editing system in any one of Claims 1-9.

Images