FR2634922A1 - METHOD FOR ROTATING AROUND AN AXIS OF AN OBJECT DISPLAYED IN 3D REPRESENTATION USING A 2D INPUT CONTROLLER SUCH AS A MOUSE - Google Patents

Abstract

A method for rotating an object displayed in three-dimensional representation on a computer-controlled video display system including a computer and a video display is disclosed. This method comprises the following steps: - display of a reference circle C on the video display, - use of a pointer control member, for example a mouse, to selectively position a pointer P, Q displayed on the video display, - indication to the computer that it must activate a control movement mode, the movement of the pointer displayed in control movement mode making it possible to rotate the displayed object around an axis determined by the direction of the control movement of the pointer and by the position of the control movement of the pointer relative to the reference circle, - in control movement mode, movement of the pointer by means of the pointer control member, and - rotation of the object displayed according to pointer movement in control movement mode.

Description

The present invention relates to a technique of emulating a controller

  3D (three-dimensional) computer input

  by a 2D (two-dimensional) computer input controller.

  Advances in computer graphics have multiplied the possibilities offered to the user. We can now display objects in three-dimensional representation, by

  example in mesh form, in full form and / or shaded.

  Although a 3D trackball input controller has already been used to directly manipulate objects displayed in three-dimensional representation, such

device is complex and expensive.

  Various techniques have been developed using 2D input controllers such as mice to manipulate

  objects displayed in three-dimensional representation.

  In a known technique, graphically displayed X, Y, and Z cursors are used which the user can adjust (for example, by means of an input controller such as a mouse) to indicate, separately, the magnitude rotation around each of the axes. Generally, only one

only slider at a time.

  Another known technique involves the menu selection of the axis around which it is desired to perform the rotation. We then move in a direction an input controller such as

  than a mouse to indicate the extent of the rotation.

  Another technique requires pressing one of three mouse or keyboard buttons to select the axis of rotation, then move the mouse

  in one direction to indicate the extent of the rotation.

  An important point to be considered with the known techniques of manipulating displayed objects represented in three-dimensional form is the absence of kinesthetic link (stimulus / response association) between the movement of the input controller and the direction of motion. rotation of the object. In other words, the movement that must be given to the input controller does not provide the sensation

  effective of a rotation of the displayed object. - -

  Another important point in the techniques using 2D input controllers for manipulating three-dimensional objects is the lack of the ability to continuously vary the axis of rotation in three-dimensional space. For example, with the graphical cursor technique, for a given rotation the axis must necessarily be one of the three orthogonal axes. Another point to consider is, in known techniques, the impossibility of rotating around an axis

  any having components along X, Y and Z ..

  For this reason, it would be interesting to have an advanced technique which allows, by means of a 2D input controller, to rotate objects displayed in three-dimensional representation while having a kinesthetic correspondence between the movement of the object.

  input controller and the rotation of the displayed object.

  It would also be advantageous to have an improved technique for rotating objects displayed in three-dimensional space around any axis. This is made possible according to the invention by a method for rotating an object displayed in three-dimensional representation on a computer-controlled display system comprising a computer and a video display using. a 2D input controller for positioning a reference indicator recognized by the computer. There is a reference circle visible to the user, and. the computer receives a signal enabling it to activate a control movement mode where the movement of the reference indicator corresponds, in this mode, to the rotation of the displayed object with respect to any axis which is determined by the management of the control movement of the reference indicator and the location of the

  control movement relative to the reference circle.

  More precisely, according to a first aspect of the invention, this method comprises the following steps: establishment of a reference circle; use of a user-controlled input controller for. selectively position a reference indicator recognized by the computer; indication to the computer that it must activate

- / 2634922

  a control movement mode, the movement of the reference indicator in control movement mode making it possible to rotate the displayed object around an axis determined by the direction of the control movement of the reference indicator and by the position of the control movement of the reference indicator relative to the reference circle; in control movement mode, moving the reference indicator by means of the input controller; and rotating the displayed object according to the

  - defined motion in control movement mode.

  According to a second aspect of the invention, the method comprises the following steps: displaying a reference circle on the video display; using a pointer controller to selectively position a pointer displayed on the video display; indication to the computer that it must activate a control movement mode, the movement of the pointer displayed in the movement mode. control device for rotating the displayed object about an axis determined by the direction of the pointer control movement and the position of the pointer control movement relative to the reference circle; in control movement mode, moving the pointer by means of the pointer control member; and rotating the displayed object according to the movement of the pointer in

control movement. -

  The advantages and features of the invention will readily become apparent to those skilled in the art upon reading. the

  detailed description below, made with reference to

attached drawings.

  Figure 1 is a block diagram. a computer system that can be used to implement

the invention that will be described.

  Figure 2 illustrates a coordinate system relating to the displayed object to be rotated in accordance with the invention. FIG. 3 schematically illustrates the movement of a displayed pointer, which is used, according to the invention,

  to produce the rotation of a displayed object.

  FIGS. 4 and 5 schematically illustrate particular cases of movements of the displayed pointer which make it possible to rotate according to the invention a displayed object. Figure 6. Diagrammatically illustrates a generalized case of movement of the pointer displayed to rotate according to

the invention a displayed object.

  FIG. 7 presents a flowchart of the various steps of the invention which make it possible to turn a

  object displayed by the movement of a displayed pointer.

  FIG. 8 presents a detailed flowchart of certain functions enabled by the flowchart of FIG. 7,

  In the detailed description that follows and in the

  various figures of the drawings, the identical elements are

  identified by the same numerical references.

  Definitions The invention that will be described relates, in general, the manipulation of a computer-displayed object represented in three-dimensional form, and it may be useful to quickly present the computer environment concerned. Fig. 1 is a generalized block diagram of a suitable computer system 10 which includes a central processing unit and memory unit 11 including, in general, a microprocessor, the associated logic circuits and memory circuits. A keyboard 13 delivers input signals to the central processing unit and memory 11, as well as a 2D input controller 15 which can be, for example, a mouse, a 2D trackball, a joystick broom, touch screen, touch pad or digitizer. Disk drives 17, which may include fixed disk drives, are used for mass storage of programs and data. The output display is a video display 19. With reference to FIG. 2, the object observed on the video display 19 may, for convenience, be referenced with respect to a system of orthogonal coordinates having its origin in the center of rotation of the object. The horizontal axis 0 is the X axis, the vertical axis is the Y axis and the Z axis is

directed to the observer.-

  For ease of explanation, the following presentation will be made in the context of a 2D input controller which is a mouse, but those skilled in the art will readily understand that the described techniques may be

  implemented with other 2D input controllers. A-

  An example of a mouse used with a computer-controlled display system is given in US-A-4,464,652, which will be considered as the teachings of the computer.

this description.

  A mouse controls the position of a mouse pointer that is displayed on the video display. The pointer is moved by moving the mouse on a flat surface in the

  direction that you want to move the pointer.

  Thus, the movement in two directions of the mouse on the flat surface is reflected on the video display by a homologous movement in two directions of the mouse pointer. Typically, a mouse has one or more. control buttons operated with fingertips. Although one can use the control buttons for different functions, such as for example the selection of a menu item pointed to by the pointer, the invention described here advantageously uses a single button mouse to

  trace the movement of the pointer on a given path.

  Specifically, the pointer at the starting point

  wanted, press the mouse button to signal to -

  the computer to activate a control movement mode, and move the mouse while keeping the button pressed. After the desired path has been drawn, release the mouse button. It is sometimes said that one

  "drag" the mouse pointer.

  The position of the mouse pointer is, typically, sampled at a predetermined rate, for example ten times per second, each sampled point defining the beginning or the end of a line segment. The path drawn by dragging the mouse pointer can therefore be considered to be formed of a series of short straight line segments connected to each other, the ends of the line segments being defined by the sampled positions of the line.

pointer of the mouse.

  Note that one can also use a predetermined key of a keyboard to activate the sliding of the

pointer of the mouse.

detailed description

  Referring now to FIG. 3, a reference circle C serves as a reference for the inputs provided by the user by means of the 2D input controller 15. In the particular case of a mouse, the circle of reference is displayed in a convenient location on the video display. For input controllers such as touch pads or scanners, where there is a match between the physical position of a physical pointer and the position of the displayed image, the reference circle could be on the

appropriate entry shelf.

  In the particular example of use with a mouse, the reference circle may surround the object to rotate, which produces the impression that the object is integral with a transparent or virtual sphere that is

  would rotate through appropriate commands.

  The center O of the circle coincides with the origin of the system of

  orthogonal coordinates of Figure 2 which, as we have

  explained above, is also the center of rotation.

  The reference circle C can be considered to represent a flattened upper hemisphere of a control ball 3D input controller, the term "upper hemisphere" corresponding to the upper half of the control ball as if it were oriented on a support surface. Of course, the reference circle is a top plan view of the flattened upper hemisphere of the trackball. As will be described in more detail, the action of moving the physical pointer or displayed, as the case may be, of a 2D input controller with respect to the reference circle is analogous to

  the rotating action of a 3D trackball.

  For example, a mouse can be used to move a mouse pointer either inside the reference circle or outside the reference circle. Dragging the mouse pointer (that is, moving the mouse by pressing its button) on or off the reference circle C causes rotation relative to the axis Z. If the mouse pointer is dragged inside the reference circle, a rotation is obtained with respect to any axes of rotation which may have components according to X, Y and Z. More precisely, for each a line segment defined by the sampled locations of the mouse pointer that has been dragged, the object displayed is rotated according to the position, length, and direction of the line segment. One of these line segments, defined by the samples of two successive positions of a mouse pointer that has been dragged, is schematically illustrated by a vector of origin P and end Q. following, the vectors from

  center O at points P and Q will be called vectors P and Q.-

  To explain how the axis is determined

  of rotation defined by an arbitrary vector D, it can be.

  it is useful to begin by explaining the particular case illustrated schematically in FIG. 4, where the vector D- has its origin at the center 0 of the reference circle and forms an angle -. with the X axis. The axis of rotation, which can be conveniently represented by the vector A, is calculated in the following way: A (x, y, z) = [-sin T cos T 0] (Equation 1) Note that the axis of rotation obtained from

  Equation (1) is limited to the XY plane.

  If we now take the particular case illustrated schematically in Figure 5, where the vector D originates at a point on the axis X which is moved in the positive direction relative to the origin and forms an angle r by relative to the X axis. The axis of rotation is obtained from equation (1), but with a rotation of X degrees-around the Y axis: DP = f (x) = f (O) (Equation 2) OP being the distance separating the center O from the circle of the origin P of the vector D, OR being the radius of the circle 11l and f (x) being an increasing monotonic function satisfying the following conditions: f (x) = 0 if x = 0 (Equation 3) f (x) = 900 if x = 1 As a special example, we can have for f (x): f (x) = 90 * x (Equation 4) The function f (x) is an interpolation function that allows the axis of rotation to extend in any plane that intersects the Y axis. Essentially, the function f (x) defines the way in which the upper hemisphere would be flattened. a 3D trackball for obtain the reference circle C as a function of the position of the axis of rotation. The function f (x) is analogous to a map projection that a cartographer would use to present

  points of the earth on a map.

- 263492-Z

  The vector A (which represents the axis of rotation) is determined in the following way: cos w 0 -sin w X (x, y, z) = [-sin T cos T 0] * 0 1 0 sin w 0 cos w (Equation 5) In the general case, schematically illustrated in FIG. 6 where P is situated at any point at an angle e with respect to the axis X, the vector D forms an angle ('8 + t) by This is equivalent, in the general case, to rotating the vector D of the particular case represented by equation (4) by 8 degrees around the Z axis. In this way, the vector A which represents the axis of rotation is obtained by modifying equation (4) to include the rotation of e degrees around the Z axis: Cs m 0 -s in e (xyz) = [-sin t cos T] i 0 - _ _ sin w 0 cos X cos 8 sin 8 0 _ _ * sin 8 cos 0 o: O- O 1i - (Equation 6) Equation (6) corresponds to Equation (5), modified so that the axis of rotation can be positioned in any way in the three-dimensional space. It will be easy to see that equation 6 is simplified by giving equations 1 and 5 for the particular cases discussed above, by zeroing the

  appropriate values of angles X and 8.

  The magnitude * of the rotation of a given vector D can be calculated from the modulus of the vector D. For example, it would be possible to calculate by multiplying the modulus of the vector D by an appropriate scale factor. However, to conform more precisely to the rotation of a 3D control ball, the magnitude of the rotation must be proportioned so that the following properties are present; (1) a complete scan of the mouse passing through the circle and passing through the center O produces a

rotation of 180.

  (2) a complete revolution of the edge of the circle (or out of it) produces a rotation of 360 about the Z axis. In the example where af (x) = 90 * x, we determined empirically that the preceding rotational properties can be satisfactorily approximated by calculating the magnitude X of the rotation, expressed in degrees, as follows: * IDI * [((1-2) X (1-IcosTi) OR l 90 (Equation 7) IDI being the length of the vector D and OR being the radius

of the reference circle.

  It will be noted that, with different interpolation functions f (x), the formula for calculating the magnitude of the rotation will be different if it is to obtain the

  rotational properties (1) and (2) above.

  For each vector D defined by the samples of two successive positions of the mouse pointer that has been dragged, calculates according to equations (6) and (7) an axis of rotation and the magnitude of the rotation. The data representing the displayed object is processed in a way-to render the rotation, and the display of the object is set to

day to show the rotation.

  Referring now to Figure 7, there is shown a generalized flow chart for implementing the aforementioned two-dimensional technique of manipulating objects displayed in three-dimensional representation. In a function block 110, the location of the mouse pointer that is dragged is determined. The axis of rotation A and the magnitude * of the rotation are, respectively, calculated from equations (5) and

(6) in a functional block 120.

  The data representing the object to be rotated is processed in a function block 130 so as to take account of the rotation, and the object which has been rotated is then displayed on the video display in a block.

functional 140.

  Referring now to FIG. 8, there is shown a more detailed flowchart of functions executed in function blocks 1-10 and 120 of FIG. 7. In a decision block 211, it is determined whether the button of the mouse.-is or not depressed. If not, the determination is repeated in decision block 211. If the mouse button is pressed, a decision block 213 is determined whether the mouse button has just been pressed (that is, ie, before the last determination just made, the mouse button was not pressed). If the mouse button has just been depressed, then P, in a function block 215, is given the current value of the location of the sampled mouse pointer. The process then returns to the block of

decision 211.

  If one did not just press the mouse button (that is, it was pressed at least since the last sampling), then Q is given in a function block 217. , the current value of the sampled position of the mouse pointer. The processing continues by calculating, in a function block 219, angles t and 0. Such a calculation is easy to perform, as a function of the positions of the points P and Q. In a functional block 221, the vector is calculated by making the difference between the

-vector P and the vector ô.

  In a functional block 223, the origin of the next vector D is initialized by assigning P the current value of Q. Lastly, in the function block 225, according to the equations (6) and (7), the axis of rotation and the magnitude of the rotation. By means of known techniques, the information relating to the axis of rotation and the magnitude of the rotation is used to update the display of the object, so as to restore the rotation. For example, many systems use a transformation matrix to map object data to display the data. In this case, this matrix could be appropriately modified to restore the rotation, and the transformed matrix would then be applied to the object data to determine the displayed data which presents

the rotation.

  A simplified version using the reference circle would be to transform the movement of the mouse pointer within the reference circle into a rotation about an axis strictly in the XY plane, the calculation being done according to the equation (1) above. The movement of the mouse on or off the reference circle would then produce a rotation around the Z axis. Although this simplified technique allows continuous rotation only around any axes located in the plane

  XY, it allows for simpler calculations.

  As mentioned above, the present invention can be used with a very wide variety of 2D input controllers. For those of these input controllers where the physical location of a physical pointer does not identify a position on the displayed image (for example in the case of mouse or 2D control balls), it is convenient to display on the video display the reference circle and a pointer. However, for 2D input controllers in which the physical location of a physical pointer identifies a location on the displayed image (for example in the case of scanners or touch pads). reference circle on the input controller, and you would not use the displayed pointer. For example, it would be possible to mark a reference circle on a scanning tablet, and the positions of the sampled points defined by: the movement of the digitizer slider would define the vectors D. Similarly, for a touch pad, the reference circle on the tablet would be marked and the sampled points of the movement resulting from the contact pressure of the user's finger or a stylet would define the vectors D. Essentially, the described invention transforms the two-dimensional motion applied to a 2D input controller in three-dimensional variables continuously varying, for example the three orthogonal components of any axis of rotation in the three-dimensional space. The technique that has been described makes it possible, by means of a 2D input controller, to directly and continuously manipulate a displayed object - in three-dimensional representation. More precisely, this technique makes it possible to have an excellent correspondence between the movement of the input controller and the resulting rotation of the object. This property can easily be seen when, for example, the reference circle is a displayed reference circle that surrounds the displayed object that one wants to rotate. The object to be rotated is superimposed on a displayed pointer, and the movement of the pointer gives the impression that the object is grasped and rotated. The advantages of the technique of the invention can also be appreciated by imagining that the circle of. reference represents a virtual sphere that surrounds the displayed object, the rotation being produced by rotating the sphere by the motion of an input controller or a motion applied to an input controller. In simple terms, we visualize what we do. The technique described here is very efficient and can be easily implemented with existing 2D input controllers. In many systems installed, the described technique can be implemented.

  implemented by installing appropriate software.

  As noted above, the described technique emulates a 3D trackball input controller by using a 2D input controller. The reference circle represents a plan view of the exposed upper part of the 3D trackball. Moving the pointer inside the reference circle is like rolling the trackball, while moving the pointer around the perimeter of the reference circle is like rotating

the ball-command.

  the emulation of a 3D control bottle, the%. The invention provides a number of features with respect to the 3D trackball. It is difficult to roll and rotate at the same time a 3D trackball because the upper hemisphere of the trackball can not be fully exposed because osn necessarily needs to place a rotation sensor at the "equator "from the trackball. In addition, the three rotation sensors of a 3D o10 trackball are orthogonally positioned relative to each other and must allow some slippage when the rotation.I the trackball is not parallel to the rolling direction of one of the given sensors. The consequence of this sliding is a possible reduction in the accuracy of measuring the rotation. In out, t. 3D ball track rotation measurement system includes important elements that are movable and-must remain accurately aligned, and that may not be

  Robust enough in hostile environments.

  With the two-dimensional technique described, we fully represent the upper hemisphere of a 3D control ball, and it is possible to simulate at the same time the rolling and pivoting. In addition, since the mechanical coupling is limited to two dimensions, the printing due to slippage is reduced. Finally, the bidimensio technique: It can be implemented with a small number of moving parts,

  which provides increased reliability and robustness.

Claims (11)

  1. A process for rotating an object. displayed in three-dimensional representation on a video display system. computer-controlled instrument comprising a computer (11-17) and a video display (19), characterized in that it comprises the following steps: - establishment of a reference circle (C), - use of a input controller. (15) actuated by the user to selectively position an indicator   reference (P, Q) recognized by the computer, -   indication to the computer that it must activate a control movement mode, the movement of the reference indicator in control movement mode making it possible to rotate the displayed object around an axis determined by the direction of the control movement. control movement of the reference indicator and by the position of the control movement of the reference indicator with respect to the reference circle, - in the control movement mode, - displacement of the reference indicator by means of the input controller, e. rotation of the displayed object according to the motion   defined in control movement mode.   The method of claim 1, wherein a movement of the reference indicator on or off the reference circle causes a rotation around it   an axis (Z) oriented towards the observer.   3. The method of claim 1, wherein a controlled movement of the reference indicator within the reference circle produces a rotation.   around any axis of the three-dimensional space.   4. The method of claim 1, wherein the step of indicating to the computer that it must activate the control movement mode comprises a step of actuating a contactor. 5. The process of claim 1, wherein the   reference circle is displayed on the video display.   The process of claim 1, wherein the   e reference is on the input controller.   7. The process of claim 1, wherein   the reference indicator is displayed on the video display.   The process of claim 1, wherein   the reference indicator is on the input controller.   9. A method for rotating an object displayed in three-dimensional representation on a computer-controlled video display system comprising a computer (11-17) and a video display (19), characterized in that it comprises the following steps : - display of a reference circle (C) on the video display, - use of a pc control unit :: .- r (15) to selectively position a pointer (E. displayed on the video display , indication to the computer that it must activate a control movement mode, the movement of the pointer displayed in control movement mode making it possible to rotate the displayed object around an axis determined by the direction of the movement of control of the pointer and by the position of the control movement of the pointer relative to the reference circle, - in control movement mode, movement of the pointer by means of the pointer control member, and - rotation of the displayed object according to the movement of   pointer in control movement mode.   The process of claim 9, wherein the   circle of reference surrounds the object displayed at 4 zre turn.   11. The method of claim 9, wherein a movement of the pointer on the reference circle out of it produces a rotation about an axis (z) oriented toward the observer. - 12. The process. of claim 9, wherein a control movement of the pointer within the reference circle produces a rotation about any axis of the three-dimensional space.   13. The process of r :. 9, in which the step of indicating to the computer that it must activate the control movement mode comprises a step of actuating a contactor. -   14. The process of claim 13, wherein   the pointer control member (15) is a mouse.   15. The method of claim 14, wherein the contactor is on the mouse.   The method of claim 14, wherein the   It contains only one button.