FIELD OF THE INVENTION

The present invention relates to methods of depicting images, and in particular to depiction of images on vehicles. More particularly, but not exclusively, the invention relates to the depiction of images of an advertising or promotional nature on vehicles.

BACKGROUND TO THE INVENTION

When images, which are typically of an advertising or promotional nature, are applied to substrates in the form of exterior surfaces of vehicles, their positioning and orientation on the vehicle will usually be chosen to be where they can be viewed as upright, by persons outside the vehicle. This presents no difficulties with generally planar, upright surfaces such as the sides of vehicles (in particular, some commercial delivery vans, buses and refrigerated trucks) and in some cases the rear end of such vehicles. However, slanted orientations of some vehicle surfaces (typically vehicle bonnets) cause images that are depicted on those surfaces, to be distorted when viewed from typical vantage points or observation angles.

By way of example, if an image such as shown in Figure 1 (which merely serves as an arbitrary example) were applied to the bonnet (engine cover or hood) of a vehicle according to the prior art, as shown in Figure 2, then the image as applied would have the same proportions as those of the original image shown in Figure 1 - although the image may be scaled to suit the available size of the substrate surface - e.g. the image may be scaled (without distortion and while retaining its proportions) so that its width fits across the vehicle bonnet. Further, the image, once applied, will follow the curvature of the vehicle bonnet, as depicted in Figure 6. As a result, when the image is viewed horizontally, e.g. from the typical vantage point of the driver of another vehicle, the image would appear distorted to the viewer as shown in Figure 3, with a perceived height of the image dramatically reduced due to the orientation of the substrate (bonnet), and the perceived width of the image while comparatively not so obviously diminished, will be further visually distorted depending on the complexity of the curvature of the vehicle bonnet. The present invention seeks to overcome the image presentation difficulties resulting from the orientations of surfaces on vehicles on which the images are applied.

SUMMARY OF THE INVENTION

According to the present invention there is provided a method of depicting an image, said method comprising: obtaining a source image in the form of an undistorted planar, two-dimensional image with dimensions that are in desired proportions relative to one another; manipulating the source image to create a manipulated image, by distorting the source image along at least one axis; and providing the manipulated image in a format suitable to be applied to a substrate in the form of an exterior surface of a vehicle, to provide an applied image; wherein the axes along which the source image is distorted is selected to compensate for an angular orientation of the substrate relative to a vantage point, such that when the applied image is viewed from the vantage point, the angular orientation of the applied image causes the visually perceived dimensions of the viewed image, to accord with the proportions between the dimensions of the source image.

The axis along which the source image is distorted may be aligned with a viewing direction, in which the applied image will be viewed from the vantage point.

The source image may be manipulated in multiple directions or along multiple axes and the extent and/or orientation of the distortion may vary across the image, to compensate for variations along the surface of the substrate, e.g. different parts of the source image may be manipulated in different directions, along different axes, and/or to different extents, to compensate for complex curvatures and/or shapes of the surface of the substrate.

The source image may be manipulated at least in part, by inverse perspective transformation (IPT). The method may include the further step of applying the manipulated image to the substrate in the form of an exterior surface of a vehicle.

The invention extends to the manipulated image or the applied image in any format, including a digital file, a two-dimensional depiction of the image, or the like and to a substrate bearing the manipulated image or the applied image.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the present invention, and to show how it may be carried to effect, the invention will now be described by way of non-limiting example, with reference to the accompanying drawings in which:

Figure 1 shows an example of a source image;

Figure 2 shows a top view of a replica of part of a typical vehicle bonnet, (which for the purposes of this specification shall now be referred to as the “vehicle bonnet”), with the source image of Figure 1 applied to the vehicle bonnet as an applied image, according to the prior art;

Figure 3 shows a front perspective view, from a desired viewpoint I vantage point, of the vehicle bonnet of Figure 2;

Figure 4 shows a top view of a vehicle bonnet with the source image of Figure 1 applied to the vehicle bonnet as an applied image according to the present invention;

Figure 5 shows a front perspective view, from a desired viewpoint I vantage point, of the vehicle bonnet of Figure 4;

Figure 6 shows a representation of projection of the image of Figure 1 , when applied to the vehicle bonnet as shown in Figures 2 and 3; and

Figure 7 shows a representation of projection of the image of Figure 1 , when applied to the vehicle bonnet as shown in Figures 4 and 5.

DETAILED DESCRIPTION OF THE DRAWINGS

Referring to the drawings, an image intended for application to a motor vehicle is generally identified by reference sign 10, but the image is also identified by suffixes, referring to the real or perceived proportions of the image. An example of a substrate in the form of an exterior surface of a vehicle such as a vehicle bonnet on which the image depicted, is generally identified by reference sign 12.

Referring to Figures 1 to 3, the application of a source image 10A as shown in Figure 1 , on a vehicle bonnet 12 using the prior art, has been described above in the background to the invention. The source image 10A is an undistorted planar, two- dimensional image with dimensions that are in desired proportions relative to one another and according to the prior art, it is applied to the bonnet 14 of the vehicle 12 in the form of an applied image 10B without any manipulation such as scaling or distortion. However, as shown in Figure 3, when the applied image 10B is viewed from a typical vantage point, the viewed image 10C is visually distorted and becomes difficult to read. In particular, the perceived height of the viewed image 10C is diminished by the slanted orientation of the bonnet surface, relative to the line of sight, while the perceived width of the viewed image 10C is not distorted in the same manner and the viewed image thus appears disproportionately wide, when compared to the source image 10A. Further, in the case of pronounced curvature and/or various orientations of parts of the surface of the vehicle bonnet 12, there are further distortions of the source image 10A in the viewed image 10C as seen in Figure 3.

Referring to Figure 4, when the source image 10A of Figure 1 needs to be applied to a substrate in the form of the vehicle bonnet 12 according to the present invention, the source image is manipulated to provide a manipulated image by distorting the source image along at least one axis 14 to compensate for an angled orientation of the bonnet, relative to a typical vantage point of an observer. The axis 14 along which the source image 10A is distorted is aligned with a viewing direction, in which the applied image will be viewed from a vantage point. If the orientation of the bonnet 12 were perpendicular to a typical line of sight from the vantage point, then no manipulation of the source image 10A were required, but in the illustrated example, the bonnet is angled and to compensate for this angular orientation, the height of the source image 10A is increased disproportionally along the axis 14 to the width of the image to provide the manipulated image.

The manipulated image is applied to the bonnet 12 in a suitable manner, e.g. by painting, printing and/or adhesive attachment such as printing the manipulated image on an adhesive sheet such as vinyl, which can be adhesively applied to the bonnet, in the form of the applied image 10D as shown in Figure 4. If necessary, an adhesive sheet to which the manipulated image has been applied, needs to be shaped to match the surface of the bonnet 12. This can usually be achieved through elasticity of the sheet, which allows it to bend and stretch as necessary, but it may be necessary in some instances to cut an adhesive sheet to allow it to follow the bonnet surface.

When the applied image 10D, is viewed from a typical vantage point, as shown in Figure 5, then the angular orientations of different parts of the applied image on the bonnet 14 causes the visually perceived dimensions of the perceived or viewed image 10E, to accord with the proportions between the dimensions of the source image 10A - i.e. the viewed image 10E is perceived as having the same proportions as the source image 10A shown in Figure 1.

Further to the distortion or elongation of the applied image 10D along the axis 14, the image is also curved (in particular, it is curved “downwards” as shown in Figure 4, to compensate for a convex shape in the centre of the vehicle bonnet 12), and the image is distorted angularly along folds 16 of the vehicle bonnet. It is anticipated that in most cases the substrates on which images will be applied on vehicles will be more complex than a single angled orientation of the substrate relative to the line of sight from a typical vantage point, and in most embodiments of the invention, source images will be manipulated with more complexity to compensate for complex curves and orientations of the substrate surfaces This could include manipulation of the source image in multiple directions or along multiple axes and the extent and/or orientation of the distortion may vary across the image, to compensate for variations along the surface of the substrate, such as complex curves. The manipulation of the source image could include inverse perspective transformation (IPT), which is described in detail in WO93/04559 and is described below. An IPT image can be generally described as a complex elongation of the image along the line of sight where the complex elongation is a result of applying a combination of desired visual image size as well as progressive stretch and key-stoning. The precise determination of this combination is dictated by the combined relationship of height and distance of the typical vantage point or viewpoint from the intended IPT image position.

Any two-dimensional object (i.e. a plane) will appear differently when viewed from a different view point or vantage point in three-dimensional space and two-dimensional representations of the appearance of the object/plane can be created using the laws of perspective. Typically, coordinates are known that describe the orientation and position of the plane (e.g. the location of each corner of a square plane relative to three orthogonal axes) and formulas are used to determine depict the plane, as perceived from the vantage point, on a two-dimensional image.

A typical example of a two-dimensional object or plane can be a square or a grid, comprising an array of adjacent squares. Each square will be perceived as a quadrilateral comprising four points joined by four straight lines. The position of each of the corners of the squares as they will be perceived from a vantage point, can be calculated and represented in a two-dimensional perspective view.

A grid of squares demarcated onto a planar substrate can be applied to a complex surface which could include curves and/or angular folds. The complex surface is viewed with the grid applied to it, the grid visually accentuates the non-planar features of the complex surface, which could otherwise be difficult for a viewer to appreciate visually.

In a view of such a grid, applied to a complex surface, from a particular viewpoint, the squares of the grid will present an array of interconnected quadrilaterals, some of which may appear to have kinked sides where the grid lines fall across folds in the complex surface, and curved sides where the grid lines fall on curved parts of the complex surface. As the viewpoint changes with respect to the surface, so the visual representation of these quadrilaterals changes.

However, for a specific viewpoint then, the visual representation can be assessed as a two-dimensional representation of the three-dimensional surface, where every visible point of the actual surface can be mapped onto a two-dimensional system (in the same way that a photograph captures a three-dimensional scene and presents it as a two-dimensional image).

If this methodology of applying a grid of squares on a complex surface is used in the present invention, as shown in Figure 7, some squares, of which one example is marked 18, fall on a flat (or a reasonably flat) plane that is simply at an orientation in space with respect to the viewer, and it will present a generally quadrilateral shape when viewed from different angles.

In some cases a grid square 20 will fall across a fold 16 in the surface and since this grid square includes two plane surfaces at different orientations with respect to the viewer, the grid square will present a “quadrilateral” with two kinks 22 in two of the sides (strictly speaking, a “hexalateral”). These kinks 22 can be joined by a straight line, demarcating a fold line of the fold 16, thus creating two quadrilaterals within that original grid square.

In some cases, a grid square 24 can present sides that are curved. If the sides of a grid square display only very slight curvature, then the curvature can be ignored, and the grid square can be resolved as a quadrilateral shape with straight sides as in 18 above. However, if the curvature is significant, each of the curved sides can be broken down into an appropriate and finite number of smaller straight lines approximating the curve, and the points of inflection between each of the straight can be connected by straight lines to corresponding points of inflection on the opposite side of the grid square so that the grid square is made up of two or more smaller quadrilaterals, each with plane surfaces at different orientations with respect to the viewer. In this way, a complex surface comprising curvatures and angular folds can be assessed as a sequence of connected grid squares, each of which comprises a plane surface at a different orientation with respect to the viewer and which may be assessed further as necessary as a sequence of smaller quadrilaterals contained within it comprising plane surfaces at different orientations with respect to the viewer.

A similar grid can be applied to the source image 10A and the portion of the source image in each grid square can be converted (distorted) to take the shape of a corresponding grid square on the complex surface where that portion of the source image is to be applied. The converted portions of the source image 10A are then combined to form the applied image 10D.

One of the methods that can be used to convert images or parts of images from source image 10A to the applied image 10D in the present invention, is best described as the Inverse Perspective Transformation (IPT) of an image, which involves carrying out a pre-distortion of a source image 10A, in accordance with the laws of perspective applicable to i) the inclined plane of the visual surface/substrate onto which the image is to be applied and ii) the specific vantage point from where it is to be viewed (viewpoint), such that the resulting viewed image 10E will appear undistorted or as a correct representation of the source image, in spite of the visual surface/substrate onto which the image is to be applied being at an inclined plane with respect to the viewpoint.

By virtue of existing IPT methodology, it is possible to transfer any information that is contained in a virtual shape on the virtual plane to that of the actual shape on the actual plane.

Any change in the inclination of the plane, although the viewpoint remains unchanged, will necessitate a different pre-distortion to achieve the same desired visual result.

Therefore, if the visual surface/substrate is made up of two or more intersecting planes of different inclinations, and given that any two planes intersect along a straight line, the parts of the desired image that fall onto the respective planes of different inclination (so defined by these lines of intersection), will each need to be subjected to a different pre-distortion so as to achieve the desired visual result across these two or more intersecting planes.

Further, the two or more intersecting planes of different inclinations making up the visual surface/substrate could be the result of angular folds and/or the presence of varying curvatures, given that any curved surface can be simplified or approximated by a number of intersecting planes.

Therefore, it is possible to analyse any composite planar surface of varying curvatures and/or angular folds as a number (or a lattice) of individual connected planes, each at a specific inclination to the viewpoint. These individual planes can be defined by any polygon of straight sides but most notably these would be squares or triangles of suitable size, so as to make up a lattice that accurately represents the composite planar surface. Typically this is referred to as a 3D Net of triangles as is produced by commercially available 3D scanning software.

It is possible to demarcate a suitably sized grid onto a planar substrate and then apply it to a complex surface comprising curves and angular folds and then to photograph it. The resulting visual image could be traced, i.e. the visual grid digitised. However, it would be preferable to have the complex surface rendered as a 3D Net, which is possible through the use of commercially available scanners and associated software. Note that typically 3D Nets employ the use of triangular shaped elements, rather than quadrilaterals, but the principle described above with reference to grid squares, applies equally to triangles or finite elements of other shapes.

The rendering of the complex surface can be carried out through the use of a camera phone to capture a sequence of photographs (or a video clip) of the complex surface from various angles and then through the process of commercially available photogrammetry applications, the complex surface is rendered as a 3D Net. Each of the points that define the corners of the squares (or triangles or other finite elements) of the grid (or lattice) are defined, by convention, by three coordinates (x,y,z) on three orthogonal axes, which defines its position in three-dimensional space. These coordinates can be the output of 3D scanning software and define the actual/real shape of each individual plane.

Each point in three-dimensional space (e.g. as defined by x,y,z coordinates) can be fixed in two-dimensional space (x,y) relative to the vantage point. Again these can be the output of 3D scanning software and define the virtual/visual shape of each individual plane.

Taking each of the individual planes in turn (i.e. each portion of the source image 10A), knowing both its actual shape, as well as its desired visual shape (i.e. the desired appearance in the viewed image 10E - which should preferably correspond to the actual shape or source image 10A), it is possible to overlay the portion of the source image 10A onto the visual shape (viewed image 10E) and then to transpose each coordinate (or pixel) of the image from the visual shape (viewed image 10E) onto its corresponding position on the actual shape of the image, corresponding to the applied image 10D.

When all the connected planes that make up the source image 10A have been treated in this way, the end result will be a visually optimised image ready for print and then, ultimately, accurate placement and application onto the complex surface/substrate in the form of the applied image 10D.

One version of the process can be summarised as:

1 . Apply a True Grid of squares (or other shapes) of known dimensions onto the complex surface where the source image 10A is to be applied - e.g. the vehicle bonnet 12.

2. View (or simulate the view of) the True Grid from the desired viewpoint or vantage point. 3. Capture the resulting view as a Perspective Grid which now follows the profile of the bonnet 12.

4. Overlay the source image 10A onto the Perspective Grid.

5. Taking each square of the Grid in turn, transfer each coordinate (or pixel) of the source image 10A from the Perspective Grid onto its corresponding coordinate on the True Grid.

6. Interpolate between the transferred coordinates (or pixels) and infill as necessary, creating an applied image 10D as output, that can be printed on an adhesive sheet (e.g. vinyl). 7. Simultaneously produce accurate layout instructions for accurate placement of the vinyl onto the vehicle bonnet 12.