CONFERENCING ENDPOINT

Field of the invention

The invention relates to a method for automatically

adjusting a camera associated with a video conferencing endpoint, and a corresponding video conferencing endpoint .

Background

Conventional videoconferencing systems comprise a number of end-points communicating real-time video, audio and/or data (often referred to as duo video) streams over and between various networks such as WAN, LAN and circuit switched networks .

In most high end video conferencing systems, high quality cameras with pan-, tilt-, and zoom capabilities are used to capture a view of the meeting room and the participants in the conference. The cameras typically have a wide field-of- view (FOV), and high mechanical pan, tilt and zooming capability. This allows for both good overview of a meeting room, and the possibility of capturing close-up images of participants and objects. The video stream from the camera is compressed and sent to one or more receiving sites in the video conference. All sites in the conference receive live video and audio from the other sites in the

conference, thus enabling real time communication with both visual and acoustic information.

During a video conference, participants at a local site often wish to share certain visual details of physical objects with the remote site. A typical example of this is; the designer (s) of a product (e.g. a shoe) wants to discuss manufacturing problems with the manufacturer which is located on another continent. In order to show details of the manufacturing defect/challenges , the manufacturer can zoom in on the product (the shoe) and point at points/areas on product while discussing how to solve the problem with the designer. In other situations, participants may want to share information only accessible on paper, like images, diagrams, drawings or even text. Today's high quality video conference cameras are certainly capable of providing close up images of such objects. However, in order to show such details of objects, the local user must manually adjust the cameras pan, tilt and zoom to capture the desired view.

Adjustments to the camera are typically done using a standard input device, such as a keypad on a remote control or a mouse by manually controlling the cameras pan, tilt and zoom. Typically a traditional IR remote control with standard push-buttons is used to adjust the camera. A standard setup is a set of four arrow keys to control the pan and tilt, and a zoom-in and zoom-out button to control the zoom.

Manually adjusting the cameras pan/tilt/zoom to capture such small details, as described above, is a tedious and time consuming process. First, a user must activate camera control by navigating through several on-screen menu' s provided by the video conference system. Secondly, when camera control is activated, a user must manually adjust the camera using the arrow keys on the remote control. This is often an iterative process of alternately adjusting the zoom and pan/tilt.

Further, even though the camera's pan-tilt mechanism include small step motors (allowing "high resolution' movement) , the video conferencing system is often

configured to move the camera in steps to spare the user from, excessive key pushing. This works as intended when the camera is in a wide FOV, however it may cause trouble when the camera is zoomed in since the steps then become quite large .

Therefore, finding the optimal camera adjustment for on known systems often require several iterations of pushing buttons on a remote control and/or an on-screen menu system, which makes it cumbersome, distractive and time- consuming .

Summary of the invention

It is an object of the present invention to provide a method for automatically adjusting a camera associated with a video conferencing endpoint, and a video conferencing endpoint, that eliminates or reduces at least some of the drawbacks described above. The features defined in the independent claims enclosed characterize this method and video conferencing endpoint.

Brief description of the drawings

The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments of the invention, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the . invention . FIG. 1 is a schematic overview of an exemplary video conferencing endpoint,

FIG. 2 is a block diagram illustrating further principles of a video conferencing endpoint,

FIG. 3 is a schematic overview a targeting device for use with a video conferencing endpoint,

FIG. 4 is a flow chart illustrating principles of a method for adjusting a camera associated with a video conferencim endpoint ,

Figure 5a and 5b are schematic views further illustrating the operation of an automatic camera adjustment feature.

Detailed description

In the following, the present invention will be discussed by describing various embodiments, and by referring to the accompanying drawings. However, people skilled in the art will realize other applications and modi ications within the scope of the invention as defined in the enclosed independent claims.

The presented invention relates to a method for

automatically adjusting the pan, tilt and zoom of one or more cameras associated with a local video conferencing endpoint, to capture a close-up view of an area/point of interest. The invention also relates to a corresponding video conferencing endpoint .

According to the present invention, a user of a video conferencing endpoint is enabled to provide a target point visible to a camera associated with the endpoint, e.g. included in the endpoint. The target point may be an optical source, placed by the user on or near a point of interest in front of the camera. Alternatively the target point may be provided by a user pointing a laser beam, or another similar optical source, at a point of interest (typically on an object) . By analyzing the pictures

captured by the camera, the method and the corresponding video conferencing method localizes the target point in the picture (s), and determines the amount of pan and tilt needed to move the camera such that the cameras view is centered on said location defined by the target point. When the amount of pan and tilt is determined, panning, tilting and zooming of the camera commences. The camera continues to zoom until the maximum zoom of the camera is reached, or, alternatively, until the user indicates that zooming should stop. The picture analysis process according to the present invention may alternatively be running continuously to automatically detect target points. However, in a preferred embodiment of the present invention, the picture analysis process is initiated by a user.

FIG. 1 is an illustration of a videoconferencing endpoint 1 according to certain teachings of the present disclosure. The videoconferencing system 1 includes at least a

videoconferencing unit 10, one or more displays 9, at least one pan/tilt/zoom enabled video camera 6, and one or more input devices 7. The videoconferencing endpoint 1 can further include one or more peripheral devices, such as a computer (either laptop or desktop) , a Digital Versatile Disc (DVD) player, etc. In one embodiment, the

videoconferencing unit 1 is a TANDBERG Codec C90, C60, MPX 6000 or MPX 3000, and the video camera 6 is a TANDBERG PrecisionHD 1080p Camera or a TANDBERG PrecisionHD 720p Camera, all products from the Assignee of the present disclosure.

The videoconferencing unit 10 is used to establish and conduct a videoconference with remote endpoints (not shown) via a network. The videoconferencing unit 10 is connected to one ore more cameras 6, one or more displays 9, one or more speakers 5, and one or more microphones 4. Depending on the implementation, the videoconferencing unit 1 can have other common components, such as an Infrared (IR) detector for receiving IR signals from a input device

(standard remote control) 7. The camera may comprise hardware, such as processing units and memory, allowing the camera to store computer programs and perform logic

operations independently of external computers. The optical sensor in the camera may be a CCD image sensor or a CMOS sensor .

Referring now to FIG. 2 , the videoconferencing endpoint 1 according to the present invention is schematically

illustrated in more detail. The videoconferencing unit 10 has a controller 200, which can include any conventional decoders/encoders, processors, and other electronic

components known in the art and used for a

videoconferencing unit. The controller 200 is coupled to an output 215 for video, an I/O interface 217 for user

interface, and a memory 220 storing functions 222. The controller 200 is also coupled to an input 216 for

receiving video from a local camera 230 and an interface 231 for controlling the local camera 230. The video output 215 is coupled to a video input of the display 9, and the I/O interface 217 receives data from an I/O device 240, such as a remote control or other device operated by a user. For example, the I/O interface 217 comprises an IR detector which receives IR signals from an I/O device 240 comprising an IR transmitter, such that the I/O device 240 can send control data to the controller 200 via said I/O interface. In other embodiments, the I/O interface 217 may comprise other wired or wireless communication means, such as Bluetooth, iFi, cable connections, etc. The controller 200 comprises a video codec 201 and a data processor 202. The video codec 201 is responsible for processing video data to be displayed by the display 9 and to be sent to remote endpoints of the videoconference . In general, the video data can include images (pictures) captured by the camera 230 of the unit 10, video from remote endpoints of the videoconference, content from a peripheral device (e.g., VCR, DVD player, computer,

document camera, etc.), and other visual data. Operation of such a video codec 201 in the context of videoconferencing is well known in the art is not described herein.

The data processor 202 is responsible for processing data for the videoconferencing unit 10. This data includes data from the camera interface 231, communication data, commands (e.g. from the I/O interface 217), data from the Target point locator function 222, videoconference information, etc. The controller 200 is also coupled to a network interface 214, such as commonly used for a

videoconferencing unit, and the network interface 214

couples to a videoconference network known in the art.

Figure 3 shows an I/O device 240 according to one exemplary embodiment of the present invention. The I/O device 240 comprises at least an optical source 304 and a controller 302 for operating said optical source 304. The optical source 304 may be any optical source detectable by an optical sensor in the camera, for example a Light Emitting Diode (LED) , Organic Light Emitting Diode (OLED) , Laser diode, laser etc. The optical source may emit optical signals having a wavelength corresponding to that of visible light or emit optical signals in the infrared wavelength range.

According to one exemplary embodiment of the present invention, the I/O device further comprises an activation operating element such as an activation button 301 for activating the optical source 304. The I/O device further comprises a second optical source 308 emitting optical signals in the infrared wavelength range. The second optical source is used to transmit commands from the I/O device to the controller 200 of the videoconference unit 10 via the I/O interface 217. The second optical source is also operated by controller 302.

According to one exemplary embodiment of the present invention, the I/O device is included in the standard remote control for operating the video conference endpoint 1.

According to one exemplary embodiment of the present invention, the I/O device is a device separate from the standard remote control for operating the video conference endpoint 1.

According to one exemplary embodiment, the two optical sources 304, 308 are one common optical source operating in the infrared wavelength range.

The controller 200 controls operation of at least some features of the videoconferencing endpoint 1 using the operational function 222 stored in memory 220. This

operational function includes a target point locator function 222. This operational function 222 is discussed in more detail later, but a general overview of the functions 222 is provided here.

The target point locator function 222 allows the

videoconferencing unit 10 to determine the location of a target point provided by a user. The target point is the optical source 304, or alternatively a point illuminated by the optical source 304. The target point locator function 222 processes one or a series of images/pictures (or sequence of images/pictures) captured by the camera 230, and determines the location of the target point within said picture. Further, the target point locator function 222 determines the displacement of the target point relative to a center point of said picture, or, alterantively, said pictures . The target point locator function 222 calculates the amount of pan and tilt necessary to place the center of the camera's field of view in the target point location. In one embodiment, the near camera 230 is a pan-tilt-zoom camera capable of, i.e. enabled for, panning, tilting, and zooming. One or more of the panning, tilting, and zooming capabilities of the local camera 230 can be accomplished by one or more mechanical actuators 402,403,405, as are used in the art for operating pan-tilt-zoom cameras of

videoconferencing units. The interface 231 is coupled to the actuators 402,403,405, and the controller 200 controls operation of the panning, tilting, and zooming capabilities of the local camera 230 using control signals via the interface 231. Actuators 402,403,405 comprise position sensing means, allowing the actuators to determine the current position of the cameras pan, tilt and zoom,

relative to a reference position. The actuators or a controller located in the base 404 of the camera 203 report the current position of the cameras pan, tilt and zoom to the controller 200 at predefined instances, e.g. at

predefined time intervals, when one of pan, tilt or zoom is performed, etc. The controller 200 can generate control signals to control the panning, tilting, and zooming of the near camera 230. Control of a pan, and zoom camera may be implemented in various ways, and one specific

implementation of controlling actuators and providing position feedback should not be limiting to the scope of the present invention.

Alternatively, the panning, tilting, and zooming

capabilities of the near camera 102 may be electronically achieved. For example, the near camera 203 may have

processing capabilities for panning, tilting, and/or zooming, and the controller 200 can control that processing using control signals via the camera interface 231.

According to one exemplary embodiment of the present invention, the controller 200, I/O interface 217 and memory 220 comprising the target point locator 222 is located in the base 404 of the camera 230. In this embodiment, the camera can perform the method according to the present invention without communication with the video conferencing unit 10.

The video conference endpoint may be an H.323 or SIP type endpoint if it is connected to an IP network, or an H.320 type endpoint if it is connected to an ISDN network. H.323 and H.320 are standards defined by the International

Telecommunications Union (ITU). When referring to the standards, reference may specifically be made to the H.323 and H.320 ITU-T recommendations that are currently in force on the priority date of the present patent application.

Figure 4 is a flowchart illustrating the method for

adjusting a camera according to one exemplary embodiment of the present invention. According to one embodiment of the present invention an I/O device 240 associated with a local video conferencing endpoint is operated by a user of the local video conferencing endpoint. As mentioned above the I/O device 240 comprises an optical source. If a user wishes to zoom the camera in to capture a close up of an object (e.g. a small object, piece of paper, image, etc) or a person, or to just center the cameras view on a point of interest, the user positions the I/O device 240 near or on the point of interest such that the optical source is at least partly visible to the local videoconference

endpoint' s camera. The procedure illustrated in figure 4 is according to one embodiment implemented by controller 200. A starting step SI is shown, but it will be appreciated that controller 200 performs many operations and therefore a starting step should be understood to be an entry point into a

subroutine, such as a subroutine used for adjusting the camera. In decision S2 a test is made as to whether an initiating signal is received from a user, indicating that a target point has been provided. The initiating signal may, e.g., be identified by the depression of an activation operating element such as the activation button 301 on the I/O device 240. If no initiating signal (e.g. depression of the activation button) is received, then a return is made to decision S2. If an inflating signal, indicating that a target point has been provided, is received in S2, the controller 200 proceeds to step S3 to determine the

location of the target point. The step S3 of determining the location of the target point comprises processing one or more consecutive images/pictures from the camera 230 to determine the location of the optical source (or point illuminated by the optical source) within the

images/pictures. Methods for locating the target point are discussed in more detail later.

In step S4, when the location of the target point has been determined, the controller determines the pan and tilt required to center the camera's field of view on the target point (or location of the target point) . This is determined by measuring the position of the target point with respect to the center of the processed picture (s), and the amount of zoom presently employed. If not already known, the controller may request the current position of the

pan/tilt/zoom mechanism from the camera 230. The current zoom used must be taken into account when calculating the amount of pan and tilt required to center the camera's view on the target point .

When the required pan and tilt has been determined, the controller instructs the camera to start panning and tilting the determined amount in step S5. The controller also instructs the camera to start zooming in step S6.

In decision S7 , a test is made as to whether the camera has finished panning and tilting (reached the point where the center of the camera's field of view coincides with the target point) . If the camera is not finished panning and tilting, then, in decision S8, a test is made as to whether an indication is received that the user wishes to stop adjusting the camera (e.g. activation button 301 is

released). If no indication is received (e.g. the

activation button 301 is not released) , then two

alternative courses of action are possible. In the first course of action (ALT1 in figure 4) a return is made to decision S7. In the second, alternative course of action (ALT2 in figure 4), the controller repeats the steps S3-S7 to correct the required pan and tilt if e.g. the user has moved the target point or to verify/correct previous calculations .

If an indication is received (e.g. the activation button301 is released), in step S9, the camera may either stop zooming and finishing the panning and tilting required to center the cameras field of view on the target point, or the camera may alternatively return to its initial

pan/tilt/zoom position as before starting step SI.

If the camera is finished panning and tilting in decision S7 , then, a test is made in decision S10 as to whether the camera has reached its maximal zoom, which may be limited by the camera's mechanical (optical) zooming capabilities. If maximal zoom is reached, the process of adjusting the camera is ended in step S12. If maximal zoom is not

reached, in decision Sll a test is made as to whether an indication is received that the user wishes to stop

adjusting the camera (e.g. the activation button 301 is released) . If an indication is not received (e.g. the activation button 301 is not released) then a return is made to decision Sll. If an indication is received (e.g. the activation button 301 is released) then the controller instructs the camera to stop zooming and the process of adjusting the camera is ended in step S12.

According to another exemplary embodiment of the present invention, the decisions in step S2, S8 and Sll are not based on whether an operating element (a button) is

depressed or released, but rather an indication by the user to start the processes of adjusting the camera or end the process of adjusting the camera. Such an indication by the user may e.g. be the user pushing the activation button 301 once to start the process and then pushing the button again to end the process. The user may also use audible or visual indications to start and stop the process, e.g. voice command, finger/arm gestures detectable by the camera, etc.

According to yet another embodiment of the invention, if the camera is not finished panning and tilting in decision S7, a return is made to decision S7 instead of proceeding to decision S8.

According to one embodiment of the present invention, the step S3 of determining the location of the target point is performed by analyzing two or more consecutive pictures captured by the camera. The camera captures images with a frame rate (the rate at which sequential frames are

captured) of N frames per second (or N Hz) , where N may be in the range 24-100. According to this embodiment the controller 302 is configured to power the optical source 304 with a pulse train, or in some other appropriate way, such that the optical source emits short pulses (e.g. 0.1-5 ms) at a frequency M. The frequency M is half the frequency of the frame rate of the camera (M=N/2) . This means that when the camera is capturing images/pictures (frames) of the optical source, every even numbered frame will comprise a lit optical source and every odd number frame will not comprise a lit optical source, or vice versa. By receiving two consecutive pictures (or frames) and using an image analysis method of subtracting one of said consecutive frames from the other frame, the result is a picture only comprising the optical source. Hence, the location of the target point (the optical source) can be determined.

According to another exemplary embodiment, other image analysis methods for detecting objects in an image may be used to localize the target point.

Figure 5a and 5b are illustrations of the operation of the automatic camera adjustment feature of the present

invention. Figure 5a is an illustration of an image/picture captured by the camera 203, where the image/picture is to be displayed on a monitor 9 at a remote and/or local endpoint 1. The picture captured by the camera shows a person sitting at a table, and a number of documents resting on the table. Assume now that the user wishes to zoom in on the documents in order to show the content of the documents to remote participants. Using a conventional system the user would have to manually adjust the pan, tilt and zoom over several iterations using a conventional remote control. However, using the present invention, the user can simply hold the I/O device 240 in front of the documents and activate the automatic camera adjustment feature by e.g. pressing the activation button. When the user e.g. presses and holds the activation button, the optical source 304 will start emitting light detectable by the camera 230. The controller 200 will then determine the location of the target point (TP) provided by the optical source, and determine the appropriate pan and tilt for the camera, and cause the camera to center its field of view on the target point (TP) . The controller is also causing the camera to zoom in, and the controller causes the camera to continue zooming in until the user indicates otherwise (e.g. releases the activation button). (As mentioned above the indication from the user for the starting and stopping of the automatic camera adjustment feature may be other than depressing and releasing the button.) The resultant picture captured by the camera is seen in figure 5b, which illustrates that the camera has been repositioned so that the target point (TP) is now in the centre of the picture (CP) . Therefore, by the simple task of positioning the I/O device 240 on or near a spatial point of interest i a scene captured by the camera, the user has caused the camera to zoom in on the point of interest with a zoom factor selected by the user.