3D BUILDING GENERALIZATION FOR DIGITAL MAP APPLICATIONS

Abstract

A digital map application enables display of a large amount of 3D buildings or 3D structures to provide enhanced display and navigation features. The 3D models (116, 216, 316) are composed from a detailed set of attributes which, when combined, portray a highly detailed visual rendering of a physical object as it exists in real life. By selectively suppressing attributes, and in appropriate cases deriving new attributes from existing data, varying degrees of the 3D model (116, 216) can be represented in lower levels of detail with reduced processing resources and to achieve a more realistic depiction. The generalization of the 3D models can be structured as a function of the distance between the 3D model and an imaginary observer datum or other suitable reference point. In one embodiment, a plurality of contemporaneous rendering zones (34, 36, 38) are established so that a 3D model (116, 216, 316) is displayed with a particular combination or set of attributes depending which rendering zone (34, 36, 38) it is in.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

This application claims priority to U.S. Provisional Patent Application No. 61/202,585 filed Mar. 16, 2009, the entire disclosure of which is hereby incorporated by reference and relied upon.

STATEMENT OF COPYRIGHTED MATERIAL

A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the official patent file or records, but otherwise reserves all copyright rights whatsoever.

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to digital maps, and more particularly toward a method for rendering three-dimensional models of real-life physical objects in a digital map.

2. Related Art

Personal navigation devices and/or map reading devices 10, like that shown for example in FIG. 1, are configured to utilize digital maps for information, route planning and navigation purposes. The navigation system 10 includes a display screen 12 or suitable graphic user interface that portrays a network of road segments 14. For the sake of clarity, it is to be understood that the term “road” or “street” refers in the broadest sense to any geometry that supports transportation—including motorized vehicle traffic, bicycle traffic, pedestrian traffic, and the like. When configured as a portable device with position determining capabilities, the navigation device 10 can be correlated to the digital map and displayed on or referenced in the images portrayed on the display screen 12. Examples of such devices include in-car navigation systems, hand-held navigation systems, some PDAs, cell phones, and the like. Alternatively, the navigation device 10 may not be configured with position determining capabilities, such as is the case with many personal computers, some PDAs and other more basic computing devices.

The navigation device 10 shown in FIG. 10 displays a bird's eye mosaic on the left-hand side of the screen 12 and a 3D (three-dimensional) simulation or rendering on the right-hand side of the screen 12. Many navigation devices 10 include or execute software programs which enable both bird's eye mosaic and 3D renderings on the display screen 12—either simultaneously as shown in FIG. 1, or alternatively. Interest in 3D rendering in connection with navigation devices 10 is growing in the marketplace. Because a navigation device display screen 12 has a limited amount of space, however, it is most efficient to construct the underlying software programs and functionality so that large amounts of data are loaded only when there are enough pixels to display the data adequately. This relates to a concept known generally as Level of Detail (LoD). LoD is a mechanism which allows systems developers to specify a data set with lower resolution to be substituted for full resolution data in appropriate circumstances. The lower resolution data set loads faster and occupies a smaller portion of the display screen. Regions may be used in connection with LoDs for this purpose. In a Region, it is possible to specify an area of the screen which an object must occupy in order to be visible. Once the projected size of the region goes outside of these limits, it is no longer visible and the Region becomes inactive. See, for example, FIG. 2 which specifies three Regions (1-3) which are effectively nested within a digital map. Typically, the larger Region 1 is associated with a coarse or low resolution. The smaller, inside Regions 2 and 3 are associated with increasingly finer Levels of Detail. Each Region, therefore, has a set LoD that specifies the projected screen size of the region in pixels that are required for the associated region to be active. Thus, as the user's viewpoint moves closer, Regions with finer LoD become active because the Region takes up more screen space. Regions with finer LoD replace the previously loaded Regions with coarser LoDs. As successive nested Regions become active, they can either accumulate data associated with each preceding Region, or replace entirely the data of the previously loaded Region.

Although the use of Regions with an associated LoD provide benefits in terms of efficient utilization of computer processing power, they have still many drawbacks particularly in the field of navigation and 3D model renderings, where it is important that the renderings simulate live, fluid motion rather than abrupt transitions and snapshots. For example, FIG. 3 illustrates, in exemplary form, a 3D view of a city area rendered at a particular LoD. This particular LoD includes not only building size and shape, but also roof colors, pediment details, and façade details such as windows, doors and other exterior features. For purposes of navigation, an image like this provides more detail than is needed and can not justify the processing resources and time required to generate the view. In other words, the average person utilizing a navigation device 10 is interested primarily in traveling to a particular destination, and particularly how to get there. A highly detailed city view like that shown in FIG. 3, while visually interesting, presents substantially more information about far-away buildings than is required to effectively assist the person in reaching their destination. This is also quite unrealistic; in real life visual details diminish with distance.

Therefore, it is to be understood that when a personal navigation device 10 is required to display a large amount of 3D buildings or 3D structures, the available memory or power of the enabling computer presents a formidable technical limitation and/or cost factor. A typical prior art approach to addressing this issue is handled by the software that is responsible for presenting the display image. As described earlier, the approach is to load only the area that is currently in the view port of the display screen 12, and to use multiple Level of Details resolution features. These techniques, however, either fail to provide an optimal Level of Detail for navigation purposes or provide too much detail such that performance is wasted processing large amounts of unnecessary data that can slow or even overload the memory of the personal navigation device 10. They also result in a non-realistic presentation of distant objects rendered with the same level of detail as near objects.

Therefore, there is a need for an improved method for generating 3D model images of physical objects, such as buildings and points of interest (POI), and presenting such 3D models in a digital map application in an efficient, optimal, and realistic manner.

SUMMARY OF THE INVENTION

This invention relates to methods and techniques for rendering three-dimensional (3D) objects on a display screen with varying degrees of detail for digital mapping applications. A digital map is provided having at least one 3D model corresponding to a physical object in reality. A plurality of attributes are associated with the 3D model which, when combined, portray on the display screen a detailed visual rendering of the physical object as it exists in real life. According to one embodiment of the invention, a plurality of contemporaneous rendering zones are established in the digital map as viewed in the display screen 12 of a navigation device 10. These contemporaneous rendering zones include at least a proximal rendering zone and a distal rendering zone. The 3D model is selectively displayed in the display screen with varying levels of attributes depending upon which rendering zone the 3D model is in. When the 3D model is displayed in the proximal rendering zone, all or most of its attributes are used in the rendering thereby creating a detailed, lifelike image of the physical object in reality. However, when the 3D model is located in the distal rendering zone, it is portrayed with a minimal number of its attributes which requires less processing resources. The attributes can be either based on stored, and/or derived from stored attributes.

In one embodiment, the invention is distinguished from prior art techniques by enabling the addition or removal of attributes, rather than changes in pixel resolution, to determine the Level of Detail (LoD) at which a particular 3D model is rendered on the display screen. The subject method is particularly well adapted for use in guiding a traveler along a predetermined route in a digital map. The display screen of the navigation device will show 3D models with varying levels of attributes depending upon their distance away from the observer datum or other suitable reference point.

The invention also contemplates a navigation device configured to display a generalized, i.e., simplified, 3D model on its display screen in which the attributes used for the rendering are derived from existing attribute data and then attached to the 3D model. The method of generalizing 3D model attributes is beneficial for display purposes, and also advantageous for data storage/processing purposes.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other features and advantages of the present invention will become more readily appreciated when considered in connection with the following detailed description and appended drawings, wherein:

FIG. 1 is an exemplary view of a portable navigation device according to one embodiment of this invention including a display screen for presenting map data information;

FIG. 2 is a schematic illustration of the prior art technique of Regions which are nested and associated with different levels of detail (LoD) which become active depending upon the zoom level of the observer;

FIG. 3 represents the view port or display screen of a prior art navigation device 10 on which a 3D rendering of buildings in a city appear at the same Level of Detail (LoD);

FIG. 4 is an exemplary view of a 3D model rendered in connection with the digital map at a lowest LoD according to this invention;

FIG. 5 is a view as in FIG. 4 showing the same 3D model rendered in a second or intermediate LoD;

FIG. 6 is a view of the same 3D model rendered in a third or full LoD;

FIG. 7 is a depiction of a building façade as used in a 3D model showing the manner in which individual tiles may be arranged to portray a high LoD rendering;

FIG. 8A depicts a cluster of 3D models rendered with attributes to provide the highest level of detail (LoD-3) and which include façades, pediment and roof textures, etc.;

FIG. 8B is a view as in FIG. 8A showing the same building objects rendered as 3D models in LoD-2 using another set of attributes, e.g., average derived colors for the façades and/or pediments generalized from the texture attributes of LoD-3;

FIG. 8C is a view showing the building objects of FIG. 8B rendered as 3D models at LoD-1 based on a set of attributes generalized from LoD-2 data or other generalized rules, and which in this case result in a noticeable change in the model shapes and footprints;

FIG. 9 is a view of a personal navigation device according to this invention having a display screen on which three contemporaneous rendering zones (LoD-1, LoD-2, LoD-3) are provided;

FIGS. 10A-C represent a sequence of images as may be portrayed on the display screen of the device shown in FIG. 9, wherein 3D models are displayed in different LoDs and wherein the LoD for a particular 3D model will change as the model moves to a different rendering zone in the display screen;

FIG. 11A is a view of a display screen for a personal navigation device according to an alternative embodiment of this invention wherein contemporaneous rendering zones extend generally parallel to a road centerline; and

FIG. 11B is a view as in FIG. 11A showing the depiction of 3D objects on the display screen located in different rendering zones and thus rendered with different LoDs.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

Referring to the figures, wherein like numerals indicate like or corresponding parts throughout the several views, this invention pertains to digital maps as used by navigation systems, as well as other map applications which may include those viewable through internet enabled computers, PDAs, cellular phones, and the like. 3D models can be rendered from numerous individual attributes which, when combined together, result in a highly detailed, realistic visual depiction of the physical object to which they correspond. However, these same 3D models can be rendered from fewer or different attributes which result in a less detailed visual depiction of the physical object, as compared with the full-attribute rendering. And still further, the same 3D models can be rendered with a minimum number or selection of attributes which result in a very basic, coarse visual depiction of the physical object. Generally, a 3D model rendered with fewer attributes requires less computing resources than one rendered with more attributes.

Attributes are features well known to those skilled in digital map fields for other (non-3D model) applications. When applied to 3D models, the attributes may for example include meta information pertaining to object position (x, y, z), object shape, pediment shape, roof detail, and façade detail. Other attributes are certainly possible. Some attributes can even be derived from given attributes. An average color can be derived e.g. from a textured image of the façade (analyzing the single or some characteristic pixels of the image) or composed from single roof elements having different colors.

FIG. 4 depicts a simplified example of a 3D model 116 as may be represented in a digital map which, in this particular case, corresponds to a multilevel building structure in real life. The model 116 is shown here in its lowest level of detail, LoD-1, wherein the only attribute represented is its block shape and position. In other words, the model 116 shows the three-dimensional exterior structure and elevation without any other attributes relating to details or colorings. As will be described subsequently, it may be desirable to group nearby or clustered objects for the purpose of rendering them as a common model 116. The same building structure is shown in FIG. 5 at a second level of detail, LoD-2, and this 3D model is generally indicated at 216. At LoD-2, the rendered 3D model 216 includes more attribute content, such as average (i.e., derived) roof color, pediment and average wall color attributes. LoD-2 may include the block and position data attributes of the 3D model 116, or may use different attributes and/or derivations to achieve placement and footprint details. Thus, the 3D model 216 at LoD-2 more closely approximates the physical object as it exists in reality, as compared to the 3D model 116 shown in FIG. 4. Therefore, it follows naturally that the processing resources needed to render the 3D model 216 are greater than those required to render the 3D model 116.

FIG. 6 is a view of the same building structure but this time the model is shown in LoD-3 and generally indicated at 316. In this LoD-3, roof texture, pediment texture and façade attributes have been included with the rendering of the building to provide a visually accurate, life-like image of the physical object as it exists in real life. It should be reiterated that not all of the attributes identified above need necessarily be stored as existing or pre-processed data in the memory of the navigation device 10. Rather, some of the attributes may be derived or calculated from other attributes either on-the-fly or as pre-processed data which are later read and put on the display 12. For example, the average color attributes (e.g., roof and façade) which, in this example, are associated with the model 216 of LoD-2 (FIG. 5), may be calculated from the respective coloring attributes associated with textures of LoD-3. Of course, many other attributes in addition to those described above may be included depending upon the circumstances.

Consistent with known teachings, an attribute (like the façade texture for example) may be composed of numerous assembled components for purpose of data compression. Thus, in the example of FIG. 7, façade tiles are shown to represent separate library elements which, in the example shown here, include six discrete library components 18-28. Of course, other rendering techniques and attribute types may be applicable, which attributes can be selectively added or removed (i.e., generalized) from the model rendering in response to the distance of the 3D model from the observer datum or other suitable reference point on the display screen 12.

The concepts of this invention enable the selective generalization of attributes used to render 3D models 116, 216, 316 of physical objects. These data can be either pre-processed and stored so that an application may simply “read” the pre-processed data and put in on the display screen 12, or the other possibility is to read the original data, calculate the additional attributes and then display it without having the results stored. Naturally, several forms in between are possible as well, so that one may pre-process some of the data and calculate the remaining attributes on-the-fly. This, of course, depends on the application and hardware preferences, memory and storage availability, CPU power, time to calculate on-the-fly, and so forth.

An appropriate storage medium is provided to store the 3D model attributes and data needed for augmenting a digital map according to these principles. Such data can be converted in different formats for the display such as, for example, in KMZ/KML files. Accordingly, maps with three-dimensional information about the buildings and structures can be delivered in different formats (shape, database, files, etc.) and then accessed by an application and further processed.

Based on a set of textured buildings 316 acquired from the full use of attributes, various actions can be executed on a single building so that one gathers additional information that can be added as new attributes or features and used to render lower resolution 3D images. These for example might include computing the representative color of the building based on LoD-3 digital texture images (e.g., façade, pediment, eaves, basements, etc.) for LoD-2 presentations, or computing the representative building height from geometric details of the building element (e.g., building body, building roof, etc.) for LoD-1 views.

FIG. 8A shows a cluster of 3D models of building objects 316 placed within the context of a digital map. In this view, all of the 3D models are rendered with the highest level of detail (LoD-3) using all or most of the available attribute data, or using the stored attribute data containing the most details of façades and textures. FIG. 8B shows these same building objects rendered as 3D models 216 using fewer attributes and/or using another set of attributes, e.g., average derived colors for the façades and/or pediments, generalized from the texture attributes of LoD-3. In this case, the level of detail is LoD-2. FIG. 8C shows these same building objects again rendered as 3D models 116 at LoD-1. LoD-1 provides 3D model renderings using a relatively few or minimal number of attribute data. The building objects rendered as 3D models at LoD-1 are based in whole or in part on a set of attributes generalized from LoD-2 (or LoD-3) data so as to provide only a very rough approximation of the objects in real life. These LoD-1 renderings can appear so generalized as to result in noticeable changes in the model shapes and footprints, for example.

Based on certain defined rules which will be apparent to those of skill in the art (functional, physical, geographical, etc.), single groups can be composed to building groups at the lower level of detail settings. This is shown, for example, in FIGS. 8C where some multiple building objects have been clustered together and rendered as a unified block 116. For the building groups 116, additional information can be gathered based on the data that was collected on single buildings as described above. These might include average building group color, average building group height, computing the average color of the roof, composed building group footprint, etc. Based on the additional calculated structures or additional attributes, a process is able to generate a new data structure and it is possible to set up different levels of detail (LoD) based on the requirements. This data structure can be stored in a database or in other formats for persistent storage. As a result, complex 3D models can be accessed by a digital map application and converted to an output format in several levels of generalization. One possible scenario could be the storage of the processed data in an Oracle database, access the data by a web service and have multiple KMZ files as output that can be visualized in Google Earth.

FIG. 9 shows an exemplary navigation system 10 having a display screen 12 according to one embodiment of this invention. The display screen 12 is arranged in this example to depict a digital map in three-dimensional form so that both the road segments and any physical objects in reality are rendered using 3D modeling techniques. As shown here, a horizon line 30 coincides with a vanishing point VP. Construction lines 32 converge on the vanishing point VP for purposes of illustration only. According to the principles of this invention, a plurality of contemporaneous rendering zones are established in the digital map as viewed in the display screen 12. These contemporaneous rendering zones include a proximal rendering zone 34, an intermediate rendering zone 36, and a distal rendering zone 38. These rendering zones 34-38 may be associated with distance from an imaginary observer which, in this particular example, is presumed to be in line with the bottom edge of the display screen 12. The lower edge (0 m mark) of the display screen 12 thus functions as a reference point from which the rendering zones are gauged. As shown by the markings at left, the proximal rendering zone 34 may span a distance (relative to the observer) from 0 to 1,000 meters; the intermediate rendering zone 36 from 1,000 to 3,000 meters; and the distal rendering zone 38 from 3,000 to 15,000 meters.

Of course, these spans are offered here for exemplary purposes only and may be adjusted to suit the particular application. Furthermore, it is not essential that an intermediate rendering zone 36 be used, as adequate functionality may be achieved with only proximal 34 and distal 38 rendering zones. Similarly, more than one intermediate rendering zone 36 may be included so that four or more rendering zones are active, each rendering models with varying Levels of Detail and attribute data. Different rendering zone geometries can be established, and the rendering zone boundaries can be dynamic rather than fixed.

3D models 116, 216, 316 that appear in the display screen 12 will be selectively rendered with varying levels of attributes (i.e., different LoDs) depending upon which rendering zone the 3D model is in. 3D models 316 displayed in the proximal rendering zone 34 will be displayed with the most attributes and corresponds to LoD-3 in the example of FIG. 6. 3D models 216 appearing in the intermediate rendering zone 36 will be presented with fewer attributes, i.e., at LoD-2. 3D models 116 located in the distal rendering zone 38 will be presented or rendered with the least number of attributes and corresponding generally to the LoD-1 as shown in FIG. 4. For 3D models that extend across two or more rendering zones, rules can be established to determine which zone will have priority.

FIGS. 10A-C represent a sequence in which the navigation device 10 is transported relative to a road 14 in reality so that the 3D models 116, 216, 316 move relative to the display screen 12. As these models move from one rendering zone to the next, attributes are either added or subtracted so that the renderings change between LoD-1, -2 and -3. Thus, a 3D model 216 appearing in FIG. 10A in the intermediate rendering zone 36 passes into the proximal rendering zone 32 when the navigation system 10 is transported forwardly a sufficient distance. As the 3D model enters the proximal rendering zone 34, additional attributes are used in its rendering so that the 3D model becomes more realistic in its appearance. Accordingly, the data processing resources of the navigation system 10 are not burdened to fully process and render models 116, 216 outside of the proximal rendering zone 34. And models 116 located in the distal rendering zone 38 receive the lowest processing attention and as such are rendered with the lowest level of detail (LoD-1).

In the example of FIGS. 9 and 10A-C, a reference point is established in relation to the digital map at the lower edge (0 m) of the display screen 12. The proximal rendering zone 34 is disposed directly adjacent this reference point and the distal rendering zone 38 is spaced farthest from this reference point. FIGS. 11A and 11B depict an alternative example wherein the reference point is selected as a road center line 40. In this example, where prime designations are added for convenience, the proximal rendering zone 34′ comprises those stretches of real estate flanking the center line 40 approximately 20 meters to each side. Specific measurements are provided for exemplary purposes only. Likewise, the intermediate rendering zones 36′ are also arranged parallel to the road center line 40, outlying the proximal rendering zones 34′. In this example, the intermediate rendering zones 36′ comprise 30 meter bands along the outer edges of the proximal rendering zone 34′. The distal rendering zone 38′ comprises those portions of the digital map which are visible in the display screen 12 and lie outside of the proximal 34′ and intermediate 36′ rendering zones. In other words, the distal rendering zones 38′ comprise those spaces lying laterally outwardly from the intermediate rendering zones 36′.

As in the preceding examples, all 3D models 316 located in the proximal rendering zone 34′ will be rendered with the highest level of detail, LoD-3. 3D models 216 residing in the intermediate rendering zones 36′ will be rendered with an intermediate level of detail, LoD-2, like that shown in FIG. 5. 3D models 116 residing in the distal rendering zone 38′ will be rendered with the lowest level of detail, LoD-1. When a 3D model crosses a rendering zone boundary, e.g., when the navigation device 10 is in motion and 3D models change position on the screen 12, the 3D model is re-render with a different number, combination and/or derivation of attributes based on its new rendering zone.

FIG. 11B provides an example where 3D models 116, 216 and 316 are located in their respective rendering zones 38′, 36′ 34′. Directional arrow 42 represents navigation instructions provided by the system 10. 3D models 316 residing directly along the routed path 42 are the only models rendered in the highest level of detail, LoD-3, because they are presumed to provide the highest degree of navigational assistance and to more closely approximate realistic viewing experience. Buildings and other physical objects spaced farther (laterally) from the road center line 40 are rendered in progressively lessening detail (LoD-2 or LoD-1) because they are less significant or less useful for navigational purposes and to more closely approximate realistic viewing experience.

The foregoing invention has been described in accordance with the relevant legal standards, thus the description is exemplary rather than limiting in nature. Variations and modifications to the disclosed embodiment may become apparent to those skilled in the art and fall within the scope of the invention.

Claims

1. A method for rendering three-dimensional (3D) objects on a display screen (12) for digital mapping applications, said method comprising the steps of:

providing a digital map having at least one 3D model corresponding to a physical object in reality;

providing a navigation device (10) having a display screen (12) and being configured to display rendered images of the 3D model;

associating a plurality of attributes with the 3D model which, when combined, portray on the display screen (12) a detailed visual rendering of the physical object as it exists in real life;

establishing a reference point in the digital map;

selectively displaying the 3D model in the display screen (12) with different attributes depending upon the distance between the 3D model and the reference point.

2. A method for rendering three-dimensional (3D) objects on a display screen (12) for digital mapping applications, said method comprising the steps of:

providing a digital map having at least first and second 3D models corresponding to two different physical objects in reality;

associating a detailed set of attributes with the first 3D model which, when combined, portray a detailed visual rendering of the corresponding physical object as it exists in real life;

associating a detailed set of attributes with the second 3D model which, when combined, portray a detailed visual rendering of the corresponding physical object as it exists in real life;

providing a navigation system having a display screen (12) capable of presenting a portion of the digital map including the first and second 3D models;

establishing a reference point in the digital map, the first 3D model being spatially closer to the reference point than the second 3D model; and

displaying the first 3D model with substantially all of its detailed set of attributes while simultaneously displaying the second 3D model with a modified set of attributes generalized from its detailed set of attributes.

3. A method for rendering three-dimensional (3D) objects on a display screen (12) for digital mapping applications, said method comprising the steps of:

providing a navigation system having a display screen (12);

providing a digital map having at least one 3D model corresponding to a physical object in reality;

associating a plurality of attributes with the 3D model which, when combined, portray on the display screen (12) a detailed visual rendering of the physical object as it exists in real life;

establishing a plurality of contemporaneous rendering zones in the digital map as viewed in the display screen (12), the contemporaneous rendering zones including a proximal rendering zone (34, 34′) and a distal rendering zone (38, 38′); and

selectively displaying the 3D model in the display screen (12) with different attributes depending upon which rendering zone the 3D model is in, wherein the 3D model is displayed with the most detailed attributes when located in the proximal rendering zone (34, 34′) and with generalized attributes when located in the distal rendering zone (38, 38′).

4. The method according to claim 1 including establishing an intermediate rendering zone spaced between the proximal and distal rendering zones, and when the 3D model is in the intermediate rendering zone displaying the 3D model on the display screen (12) with more generalized attributes than when located in the proximal rendering zone and less generalized attributes than when located in the distal rendering zone.

5. The method according to claim 1, further including establishing a reference point in relation to the digital map, the proximal rendering zone disposed adjacent the reference point and the distal rendering zone spaced farthest from the reference point.

6. The method according to claim 1, wherein said step of establishing a plurality of rendering zones includes arranging the rendering zones generally parallel to the road centerline.

7. The method according to claim 1, further including the step of rendering at least two adjacent 3D models as a single building group at a lower level of detail setting.

8. The method according to claim 1, wherein the attributes of the 3D model include object shape, roof shape, average color, and façade detail.

9. The method according to claim 1, further including the step of deriving an average façade color attribute from a façade texture attribute.

10. The method according to claim 1, further including the step of deriving an average roof color attribute from a roof texture attribute.

11. The method according to claim 1 further including the step of moving the 3D model relative to the reference point in the screen, and changing the attributes displayed with the 3D model if the 3D model moves to a different rendering zone.

12. The method according to claim 1, wherein said step of establishing a reference point includes designating a centerline of the road segment as the reference point.

13. A navigation-capable device (10) configured to display a driving route on the display screen (12) according to claim 1.

14. A storage medium used to store 3D model attributes for augmenting a digital map according to claim 1.