The present disclosure relates to methods and systems for combining images of computer users with computer interfaces. More particularly, the present disclosure provides methods and systems for controlling a computer using a video image and for combining the video image with a computer desktop.
In recent years, there has been increasing emphasis on developing teleworking applications and the supporting infrastructure. Teleworking generally includes telecommuting (the ability of an employee to work from home) and distance education.
The benefits of telecommuting can be felt in both urban and rural areas. Telecommuters in urban areas can avoid commuting to the office each workday, thereby reducing stress on the employee, increasing the amount of time available for work, and reducing automobile emissions. In rural areas, employees can work for employers located in urban areas without needing to commute or relocate. Rural areas reap the economic benefits of having a higher-paid workforce without the burden of building and maintaining a transportation infrastructure. Distance education provides similar benefits by permitting students to have some flexibility in course scheduling and providing rural areas access to educational resources.
Where teleworking applications fall short is in their support for collaborative development efforts. For example, recent trends in software development involve paired programming, where programmers work side-by-side at the same computer terminal developing the same portion of code. Programmers working in pairs are believed to produce more lines of code with fewer errors than programmers working alone. Paired programming involves a high level of verbal and non-verbal interaction between the programmers. Many existing teleworking applications support some level of interaction between participants, but typically require participants to divide their attention between different portions of a display device to view the person speaking and the subject matter being discussed. In some cases, special networking hardware is required. In addition, control of a single pointing device is often shared between collaborating users, which can impede the flow of communication since one user may need to surrender control of the pointing device to permit the other user to visually indicate the subject being discussed. Thus, existing computer interfaces that sectionalize a computer display between video images of users and the application being discussed are unsuitable for paired programming.
Another area in which existing computer interfaces fall short is education. Computer-based presentations are becoming increasingly common, both in business settings and educational settings. During a presentation, the presenter may wish to visually refer to a portion of a presentation slide, typically using a pointer stick or laser pointer. To align the pointer with the desired object on the computer interface, the presenter may need to move away from the computer being used to control the presentation slides and then back to advance the presentation to the next slide. The back and forth movement of the presenter may be distracting to the viewers and may interrupt the flow of the presentation. The members of the audience may have to divide their attention between the slide being presented and the presenter, which may detract from the quality of the presentation. In addition, depending on the size and position of the display, the presenter may not be able to satisfactorily indicate the portion of the presentation slide being referenced.
Accordingly, there is a need to provide methods and systems for controlling a computer using a video image and for combining the video image with a displayed computer desktop image.
In accordance with one aspect of the present disclosure, a method for controlling a computer using a video image is provided. According to this method, a video stream is captured. The video stream is made up of a plurality of video frames. At least some of the video frames are analyzed to determine a location of an object. The location of the object may be used to control one or more programs being executed by the computer. The video stream is combined with a user interface stream generated by the computer operating system, thereby forming a composite video stream. The composite video stream is displayed using a display device.
In accordance with another aspect of the present disclosure, a method for combining a video image of a user with a computer desktop interface is provided. According to this method, a video stream containing a live image of a computer user is captured. The video stream is transparently combined with an image of a computer desktop. The combined image is then displayed. The combined image includes a transparent or partially transparent image of the user and an image of the computer desktop. The user can indicate objects on the desktop to a viewer of the display using the user's image on the desktop.
As used herein, the terms “desktop” and “desktop interface” are intended to refer to a graphical user interface that allows control of programs executing on the computer. Neither of these terms is intended to be limited to a specific computer operating system.
The methods described herein for controlling a computer using a video image and for combining the video image with a computer desktop may be implemented in hardware, software, firmware, or any combination thereof. In one exemplary implementation, the methods described herein may be implemented as computer executable instructions embodied in a computer readable medium. Exemplary computer-readable media suitable for use with the implementations described herein include disk storage devices, chip memory devices, and downloadable electrical signals that carry computer-executable instructions.
The present disclosure provides systems and methods for creating a novel user interface that supports not only single-user interactions with a personal computer, but also close pair collaboration, such as that found in distributed pair programming. In one implementation, the methods and 30 systems may be Implemented using a personal computer.
Console 104 may be of any commercially available or custom architecture and may be based on a variety of microprocessor architectures, including an Intel-style processor, a Motorola-style processor, MIPS, and others. Console 104 may include supporting memory, such as RAM and ROM, and storage devices, including magnetic disk drives and optical drives. Console 104 may also include an operating system that provides interface functionality between software applications and the underlying hardware. In one implementation, an Apple Macintosh running MacOS X 10.2 was used.
Console 104 may also include interface hardware, such as a graphics interface card, to support display device 102. The graphics interface card preferably provides standard 3D graphics capabilities and may interface with application software using the OpenGL standard. In one implementation, an nVidia GeForce4 graphics interface card was used. One feature that is preferably provided by the graphics card is the ability to transparently combine images. This feature will be described in further detail below.
Camera 106 interfaces to console 104 using a suitable high speed interface, such as IEEE 1394, also referred to as FireWire. Camera 106 is preferably capable of providing a video signal at 30 frames per second for best performance. Stored video, such as from a digital versatile disk (DVD), may be used to produce a video stream in place of or in addition to the camera 106. In one implementation, an OrangeMicro iBot camera was used to produce color video images of user 108 at 30 frames per second with a resolution of 640 pixels by 480 pixels.
Although the embodiment illustrated in
As shown in
In addition to the display aspect, the present disclosure may also include a control aspect. For example, user 108 may make predetermined movements of finger 110 to control one or more programs executing on computer 100. In one implementation, the user may train computer 100 to recognize certain movements as mouse clicks or other events to allow user 108 to interact with application programs.
In accordance with one aspect of the disclosure, video intercept process 204 forwards the video stream to a custom video analysis process 206. The video analysis process 206 provides analysis techniques to extract the position of objects in the video frame. In particular, the coordinates of an object of interest, such as a user's fingertip, are determined and passed to the user interface of the computer, shown in
The coordinates of the object of interest may be determined by converting each frame of the video stream into a two-color image by applying a filter. The filter may pass a specific color and block others. The resulting image shows a concentration of pixels corresponding to the location of the object of interest. If the image contains more than one concentration of pixels, the largest concentration of pixels closest to the previous location of the object of interest may be selected to be the object of interest. The center of the concentration is determined and used as the location of the object of interest.
Other algorithms for object detection and tracking may be used in video analysis process 206, such as edge detection or motion detection. An example algorithm for edge detection uses image analysis to determine the gradient of a greyscale colorspace image to find the most likely edges of objects. An object may then be searched for by looking for particular shapes or sizes of objects and thereby determining their placement in the image.
Motion detection algorithms detect objects in motion in a video stream by detecting differences between two subsequent video frames. The areas of difference correspond to objects in the video field of view that have moved. In a system such as the video analysis process 206, this can be used to find frequently moving objects, such as fingertips or other object of interest that a user is using to direct the video, against a nominally non-moving background.
Other approaches in video analysis process 206 may combine algorithms into new discovery techniques, such as using color filtering to provide a sample set of possible objects of interest and edge detection to further refine the set into the specific objects requested.
The location of the object of interest is passed to the mouse driver process 208, for example as a coordinate pair. The mouse driver process 208 translates the coordinate pair into a format understandable by the application 210. It should be appreciated that the term “mouse” is used generically to describe a user input device and may include other user input devices, such as a trackball, joystick, or tablet. In
Application 210 may be the computer operating system or an application running on the computer. Based on the type of mouse events reported to the application 210, the application 210 may update or change the information that is displayed on computer display device 102. For example, a window containing an application may be opened, closed, or resized and this change is displayed on display device 102. Application 210 may forward the updated display information to application stream process 212. Application stream process 212 may be provided as part of the computer operating system to provide a uniform interface for an application to update its appearance on the computer display 102. Application stream process 212 acts as an input to the transparency process 214, which may alter properties of the application stream. The output from transparency process 214 is forwarded to the compositing process 216.
Video analysis process 206 forwards the intercepted video stream to video altering process 218. Video altering process 218 may incorporate various real time filters and effects to manipulate the video stream. For example, an animated overlay layer may be added to the video stream to mark-up archived content. An edge detection filter may also be used to create a minimally intrusive line-drawing effect for the feedback video, which may influence the level of transparency of the video stream that is set by transparency process 214. Video altering process 218 forwards the altered video stream to the visual feedback process 220. Visual feedback process 220 may perform additional image manipulations to provide feedback to the user with regard to the position of the pointer. The manipulations may include horizontally reversing the images of the video stream to produce a mirror image to provide meaningful feedback to the user concerning the location of his hand relative to the desired pointer location on the desktop. The altered video stream is forwarded to transparency process 214, which may change the display properties of the video stream. The video stream is made up of a series of video frames and an alpha channel. The value of the alpha channel determines the level of transparency of the video frame images. The OpenGL standard provides an interface for changing, among other things, the alpha channel of the video stream. Transparency process 214 forwards the video stream to compositing process 216.
Compositing process 216 combines the video stream and the application stream into a single output stream which is displayed on display device 102. Compositing process 216 takes advantage of the powerful image process capabilities of 3D graphics interface cards. The video stream and the application stream images are combined to form a single video stream that is forwarded to the screen buffer and displayed on display device 102 for viewing by the user. The transparency of each stream, which is set by the respective transparency process 214, determines the level of opacity of each stream.
Referring to
In accordance with one aspect of the present disclosure, a custom video analysis routine, TrackerLib 408, is implemented as a QuickTime™ API application. TrackerLib 408 provides analysis techniques to extract positions of objects in the video frame. In particular, the coordinates of the user's fingertip are determined and passed to the user interface (Ul) of the computer, thereby acting like a human interface (Hl) device. The output of TrackerLib 408 is processed by the HI device layer 410 in a manner similar to a traditional external input device, such as a mouse ortrackball. As described above, “click” events may be generated by gestures of the user, for example by temporarily obscuring the user's fingertip. In one embodiment, obscuring the user's fingertip for less than 0.5 seconds may be interpreted by TrackerLib as a “single click”. Obscuring the user's fingertip for more than 0.5 seconds but less than 1 second may be interpreted as a “double click”. Drag events, which involve clicking on an object and dragging it from a first location on the desktop to a second location, may be generated by obscuring the user's fingertip, moving the fingertip from a first location to a second location, and un-obscuring the fingertip. The first and second locations are the endpoints of the drag event.
TrackerLib 408 uses positional and object boundary information to alter the video stream for visual feedback to the user. In the present embodiment, various real-time filters and effects 412 native to the operating system are used to perform the desired image manipulations. For example, QuickTime™ Sprites 414 may be used to mark-up archived content. Sprites are an animated overlay layer that may be used for per-object visual tracking feedback and may respond to various mouse events. Edge detection filters may be used to create a minimally intrusive line-drawing effect for the feedback video.
The Quartz Extreme layer 416 combines the video and Ul streams into a series of OpenGL 418 textures with appropriate alpha channels. The textures are composited by the accelerated video hardware's 420 3D OpenGL pipeline and sent to the display device 422. It should be appreciated that applications other than Quartz Extreme and OpenGL may be used to provide transparency control of the video streams.
In the examples described above, a single video stream is combined with a desktop application stream and composited to form an output stream that is displayed by a display device. It should be appreciated, however, that the method described in
Collaborative desktop applications currently exist that permit multiple users to control their own mouse pointer on a shared desktop. To accommodate such applications, the method depicted in
In
As previously noted, the methods and systems described herein provide an effective technique for a single user to interact with a personal computer or other computing device. These methods and systems are also effective in applications and devices where two users or more are collaborating on a task or otherwise communicating synchronously.
In one implementation, the composite display described above may be implemented by making the video image of the user transparent and placing the video image of the user on top of the desktop contents to let the desktop contents show through. In an alternate implementation the video image of the user may be placed behind the desktop contents so that the desktop contents are in full view with a faint image of the user under the desktop contents. There may be some applications for which the video image of the user may be composited with other video streams such that portions of some streams may be obscured by the video image of the user and others may obscure portions of the video of the user. Video images may be generated live from a camera or other real-time capture device or may come from a stored video source, such as a movie file or data repository. In one implementation, a live video image of the user may be displayed simultaneously with stored video content, such as a movie, so that the user can interact with the movie.
According to yet another feature of the present disclosure, the level of transparency of the user and/or desktop image may be set dynamically. During use, the user may change the video image from nearly or fully transparent image (where the user's image is not visible or is very faint, to emphasize the desktop contents) to a nearly or fully opaque image (where the user image is dominant and fully or nearly obliterates the desktop contents) to emphasize the video information and communication via the user's image. This dynamic setting may be implemented with explicit interface software controls, such as sliders, buttons, etc., in the windowing software, by hardware devices, or by image recognition of hand, arm, or face motions, or other video image content.
The methods and systems described herein can be applied to personal computer gaming, electronic games on platforms other than a personal computer, such as a game console, arcade platform, game engines running remotely over the Internet or other network, or custom game processes embedded in other products. For example, the video image of a user may be transparently combined with a game display, and the user may be able to control objects in the game display using the methods described herein.
The methods and systems described herein will function with different camera angles and locations other than directly in front of the user. It may be appropriate for different applications or usage contexts to have the camera closer or further away, above or below the level-plane of the user's eyes or at a point distant from the user to accommodate angles that provide better ease of arm motion, pointing, or user interactions.
The methods and systems described herein allow a single user to control a computer operating system without the aid of a traditional pointing device, such as a mouse or trackball. The single user embodiment may be particularly useful when access to a traditional pointing device is not convenient, such as during a presentation in a lecture hall.
The methods and systems described herein also allow multiple users to control a single computer operating system. The methods described may be combined with the networking capabilities of modern personal computers to provide a video stream from remote locations, for example to support teleworking and distance education applications.
Applications of the methods and systems described herein, in addition to those described above, include video conferencing in which multiple users may desire to transparently register their images on the same desktop and/or control the same desktop. Another exemplary application of the methods and systems described herein includes scientific visualization, 3-D graphics models, virtual reality environments, or any other material in which the image displayed is controlled by a mouse pointer. For example, in a virtual reality environment, instead of using a mechanical mouse to navigate through an image, the user may use the tracked video image of the user's finger to navigate the virtual reality environment.
In yet another application, the methods and systems described herein may be used to drive any external devices that can be driven by a computer interface. For example, telescopes include software interfaces that allow the telescopes to be driven to view particular objects. Using the methods and systems described herein, a user may simply point his finger at an object that the user desires the telescope to view and the resulting image of the user's finger may interface with the telescope control software to point the telescope at the particular object.
In yet another application, the methods and systems described herein may implement an all-video desktop. In order to implement an all-video desktop, the methods and systems described herein may be used to track additional objects other than the user's fingers. For example, the user's face, and/or icons on the desktop may be tracked in the video frame.
In yet another application, the methods and systems described herein may be used to facilitate control of a computer by handicapped users. For example, for visually handicapped users, audio signals or tactile feedback may be provided to the user as the pointer is tracked to indicate desktop objects being manipulated by the pointer.
Yet another application for the methods and systems described herein is gesture based web browsing. For example, the application being controlled by one or more users using the methods and systems described herein may be a web browser. Just as a conventional web browser may be controlled using mouse click events, the methods and systems described herein may be used to generate such events and allow users to control web browsers using a convenient interface. In collaborative web browsing, video images of multiple users may be transparently displayed with the same web browser and each user may point to or control interfaces associated with the web browser using the respective users video image. Such an application is particularly important for collaborative research where the research is being performed via Internet web pages.
It will be understood that various details of the present disclosure may be changed without departing from the scope of the present disclosure. Furthermore, the foregoing description is for the purpose of illustration only, and not for the purpose of limitation, as the present disclosure is defined by the claims as set forth hereinafter.
This application claims the benefit of U.S. Provisional Patent Application Ser. No. 60/486,516, filed Jul. 11, 2003, the disclosure of which is incorporated herein by reference in its entirety.
This invention was made with Government support under Grant No. R82-795901-3 awarded by the U.S. Environmental Protection Agency. The Government has certain rights in the invention.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/US04/22268 | 7/9/2004 | WO | 5/30/2006 |
Number | Date | Country | |
---|---|---|---|
60486516 | Jul 2003 | US |