U.S. patent application number 10/060353 was filed with the patent office on 2002-10-03 for image-based digital representation of a scenery.
Invention is credited to Niemi, Sami, Persson, Mikael.
Application Number | 20020141655 10/060353 |
Document ID | / |
Family ID | 20282823 |
Filed Date | 2002-10-03 |
United States Patent
Application |
20020141655 |
Kind Code |
A1 |
Niemi, Sami ; et
al. |
October 3, 2002 |
Image-based digital representation of a scenery
Abstract
An image-based digital representation of a scenery for visually
presenting the scenery is stored in an image information storing
structure, which comprises vertices, each representing a digital
image, and edges binding together a first vertex with a second
vertex. Each edge represents information of a transition between a
first image in the first vertex and a second image in the second
vertex, the first and the second image depicting different views of
the scenery, the first image having a first area and the second
image having a second area, wherein the depictions of the first
area substantially correspond to the depictions of the second area.
The transition information defines a manipulation of at least one
of the first and the second images, said manipulation comprising a
computation of at least one transition image, the transition image
depicting a new view of the scenery depicted in the first and
second images. Also, a method for presentation of a navigation in a
scenery, a method for presentation of images, and a device, a
method, and a computer program for creating an image information
storing structure is disclosed.
Inventors: |
Niemi, Sami; (Amiralsgatan,
SE) ; Persson, Mikael; (Kulgranden, SE) |
Correspondence
Address: |
Finnegan, Henderson, Farabow,
Garrett & Dunner, L.L.P.
1300 I Street, N.W.
Washington
DC
20005-3315
US
|
Family ID: |
20282823 |
Appl. No.: |
10/060353 |
Filed: |
February 1, 2002 |
Current U.S.
Class: |
382/276 |
Current CPC
Class: |
G09G 2370/027 20130101;
G09G 2340/045 20130101; G06F 3/14 20130101; G09G 2340/0407
20130101; G06T 3/403 20130101 |
Class at
Publication: |
382/276 |
International
Class: |
G06K 009/36 |
Foreign Application Data
Date |
Code |
Application Number |
Feb 2, 2001 |
SE |
0100325-0 |
Claims
1. An image-based digital representation of a scenery for visually
presenting the scenery, the representation being stored in an image
information storing structure comprising vertices, each
representing a digital image, and edges binding together a first
vertex with a second vertex, each edge representing information of
a transition between a first image in the first vertex and a second
image in the second vertex, the first and the second image
depicting different views of the scenery, the first image having a
first area and the second image having a second area, wherein the
depictions of the first area substantially correspond to the
depictions of the second area, said transition information defining
a manipulation of at least one of the first and the second images,
said manipulation comprising a computation of at least one
transition image, the transition image depicting a new view of the
scenery depicted in the first and second images.
2. The image-based digital representation according to claim 1,
wherein the manipulation comprise a computation of several
transition images depicting different views of the scenery.
3. The image-based digital representation according to claim 1,
wherein the transition information defines a manipulation for
altering a presentation of a first image, the depictions of which
comprising a planar surface from a first viewing point, to a
presentation of a second image, the depictions of which comprising
said planar surface from a second viewing point.
4. The image-based digital representation according to claim 1,
wherein the transition information defines a manipulation for
altering a presentation of a first image depicting a scenery from a
viewing point in a first viewing angle to a presentation of a
second image depicting the scenery from said viewing point in a
second viewing angle.
5. The image-based digital representation according to claim 1,
wherein the transition information defines a manipulation for
altering a presentation of a first image depicting a scenery from a
first viewing point in a viewing angle to a presentation of a
second image depicting the scenery from a second viewing point in
said viewing angle.
6. The image-based digital representation according to claim 1,
wherein the manipulation comprises a projective transformation.
7. The image-based digital representation according to claim 1,
wherein said manipulation is based on the information in the first
and second areas.
8. The image-based digital representation according to claim 1,
wherein said manipulation uses information of correspondence points
in the first and the second areas.
9. The image-based digital representation according to claim 8,
wherein the transition information defines a function of the
progress of the transition for the movement of the correspondence
points from a position in the first image to another position in
the second image.
10. The image-based digital representation according to claim 8,
wherein the transition information further comprises a point set,
which define the positions of the correspondence points in a
specific transition image.
11. The image-based digital representation according to claim 1,
wherein a point in the transition image is computable by blending a
mapping of a point in the first image onto said point in the
transition image with a mapping of a point in the second image onto
said point in the transition image.
12. The image-based digital representation according to claim 11,
wherein the transition information defines weights for the
blending, said weights defining the influence of each point in the
first and second images for calculation of the transition
image.
13. The image-based digital representation according to claim 12,
wherein the weights differ within the first and second image in
accordance with a predefined function.
14. The image-based digital representation according to claim 12,
wherein the weights of the first and the second images varies as a
function of the progress of the transition for calculation of
different transition images.
15. The image-based digital representation according to claim 1,
wherein the manipulation uses at least one additional image for
computation of the transition image, the first, the second and the
additional images depicting different views of the scenery, the
first image having a first area, the second image having a second
area, and the additional image having a third area, wherein the
depictions of the first area substantially correspond to the
depictions of the second area and the third area.
16. The image-based digital representation according to claim
l,wherein the transition information defines a function for
altering a presentation of at least a part of the first image to a
presentation of at least a part of the second image via a
presentation of the at least one transition image.
17. The image-based digital representation according to claim 16,
wherein the transition is reversible, whereby the transition is
usable for altering a presentation of at least a part of the second
image to a presentation of at least a part of the first image.
18. The image-based digital representation according to claim 1,
wherein the vertex representing the first image further comprises a
link to another vertex via at least one specified edge for
transition of a presentation of the first image to a presentation
of the image represented by the other vertex.
19. The image-based digital representation according to claim 18,
wherein the link is coupled to the first area.
20. The image-based digital representation according to claim 1,
wherein the first and second areas are predefined.
21. The image-based digital representation according to claim 1,
wherein the transition is controllable by input from a user.
22. A method for presentation of a navigation in a scenery
represented by digital images, said method comprising the steps of:
accessing a first and a second digital image, the first and the
second image depicting different views of the scenery, the first
image having a first area and the second image having a second
area, wherein the depictions of the first area substantially
correspond to the depictions of the second area, accessing
information of a transition between the first image and the second
image, said transition defining a manipulation of at least one of
the first and the second images, determining at least one
transition image using the defined manipulation, said transition
image depicting a new view of the scenery depicted in the first and
second images, performing a visual navigation in the scenery by
replacing a presentation of an image depicting a view of the
scenery by a presentation of another image depicting a different
view of the scenery.
23. The method according to claim 22, wherein a user controls what
image is to be presented.
24. The method according to claim 22, wherein an animation is
presented, said animation forming a specified path for altering the
presentation of an image to the presentation of another image.
25. The method according to any one of claims 24, wherein the
presentation of the animation is controllable by a user.
26. The method according to any one of the claims 22, further
comprising the step of downloading the first and the second images
to a presentation device.
27. The method according to any one of claims 26, wherein
calculations of the manipulation is performed at the presentation
device.
28. The method according to claim 26, further comprising the step
of enabling a transition from a first image to a second image, when
both images have been downloaded to the presentation device.
29. The method according to claim 22, wherein the steps of
accessing the first and the second images and the transition
comprise the step of accessing a computer program, which accesses
the image information storing structure.
30. An image information storing structure for altering a
presentation of a first image so that a transition is made to a
presentation of a second image, said image information storing
structure being accessible by a presentation device and comprising:
vertices, each representing an image, and edges binding together a
first vertex with a second vertex, each edge representing
information of a transition from a first image in the first vertex
to a second image in the second vertex, the first image having a
first area and the second image having a second area, information
in the first area substantially corresponding to information in the
second area, said transition information defining a manipulation of
at least one of the first and the second images, said manipulation
forming a function for altering a presentation of the first image
so that a transfer is made to a presentation of the second image
via a presentation of at least one transition image.
31. An image information storing structure for altering a
presentation of a first image so that a transition is made to a
presentation of a second image, said image information storing
structure being accessible by a presentation device and comprising:
a first image and a second image, the first image having a first
area and the second image having a second area, information in the
first area substantially corresponding to information in the second
area, information of a transition from a first image to a second
image, said transition information defining a manipulation of at
least one of the first and the second images, said manipulation
forming a function for altering a presentation of the first image
so that a transfer is made to a presentation of the second image
via a presentation of at least one transition image.
32. A method for presentation of digital images, said method
comprising the steps of: accessing a first and a second stored
digital image, said first image having a first area and said second
image having a second area, information in the first area
substantially corresponding to information in the second area,
accessing information of a defined transition from the first to the
second image, presenting the first image on a presentation device,
performing the transition from the first image to the second image
by calculating a manipulation of at least one of the first and the
second images, presenting at least one transition image on the
presentation device, said transition image being created by the
manipulation, presenting the second image on the presentation
device to complete the transition of presentation of the first
image to presentation of the second image.
33. A device for creating an image-based digital representation of
a scenery for visually presenting the scenery, said device
comprising means for obtaining a first and a second digital image,
the first and the second image depicting different views of the
scenery, means for defining a first area of the first image and a
second area of the second image, wherein the depictions of the
first area substantially correspond to the depictions of the second
area, means for defining information of a transition between the
first image and the second image, said transition information
defining a manipulation of at least one of the first and the second
images, said manipulation comprising a computation of at least one
transition image, the transition image depicting a new view of the
scenery depicted in the first and second images, means for storing
the first and the second images as two vertices in a graph of an
image information storing structure and the defined transition
information between the first and the second image as an edge
binding together said two vertices in the graph of the image
information storing structure.
34. The device according to claim 33, further comprising means for
presenting an overview of the graph of the image information
storing structure.
35. The device according to claim 33, further comprising means for
defining correspondence points of the first and the second images
for use in the manipulation.
36. The device according to claim 35, further comprising means for
presenting guiding lines for the definition of correspondence
points.
37. The device according to claim 35, further comprising means for
presenting a magnified view of a part of an image, in which a
correspondence point is to be defined.
38. The device according to claim 37, further comprising means for
scaling down the obtained images in resolution.
39. The device according to claim 37, wherein the means for
presenting a magnified view uses the obtained image, which is in
high resolution.
40. The device according to any one of claims 33, further
comprising means for presenting the defined transition.
41. The device according to claim 40, further comprising means for
halting a presentation of the defined transition at an arbitrary
point.
42. A method for creating an image-based digital representation of
a scenery for visually presenting the scenery, said method
comprising the steps of: obtaining a first and a second digital
image, the first and the second image depicting different views of
the scenery, defining a first area of the first image and a second
area of the second image, wherein the depictions of the first area
substantially correspond to the depictions of the second area,
defining information of a transition between the first image and
the second image, said transition information defining a
manipulation of at least one of the first and the second images,
said manipulation comprising a computation of at least one
transition image, the transition image depicting a new view of the
scenery depicted in the first and second images, storing the first
and the second images as two vertices in a graph of an image
information storing structure and the defined transition
information between the first and the second image as an edge
binding together said two vertices in the graph of the image
information storing structure.
43. The method according to claim 42, further comprising the step
of presenting an overview of the graph of the image information
storing structure.
44. The method according to claim 42, further comprising the step
of defining correspondence points of the first and the second
images for use in the manipulation.
45. The method according to claim 44, further comprising the step
of presenting guiding lines for the definition of correspondence
points.
46. The method according to claim 43, further comprising the step
of presenting a magnified view of a part of an image, in which a
correspondence point is to be defined.
47. The method according to claim 42, further comprising the step
of presenting the defined transition.
48. The method according to claim 47, further comprising the step
of halting a presentation of the defined transition at an arbitrary
point.
49. A computer program for creating an image-based digital
representation of a scenery for visually presenting the scenery,
said computer program comprising code for obtaining a first and a
second digital image, the first and the second image depicting
different views of the scenery, code for defining a first area of
the first image and a second area of the second image, wherein the
depictions of the first area substantially correspond to the
depictions of the second area, code for defining information of a
transition between the first image and the second image, said
transition information defining a manipulation of at least one of
the first and the second images, said manipulation comprising a
computation of at least one transition image, the transition image
depicting a new view of the scenery depicted in the first and
second images, code for storing the first and the second images as
two vertices in a graph of an image information storing structure
and the defined transition information between the first and the
second image as an edge binding together said two vertices in the
graph of the image information storing structure.
50. The computer program according to claim 49, further comprising
code for presenting an overview of the graph of the image
information storing structure.
51. The computer program according to claim 49, further comprising
code for defining correspondence points of the first and the second
images for use in the manipulation.
52. The computer program according to claim 51, further comprising
code for presenting guiding lines for the definition of
correspondence points.
53. The computer program according to claim 51, further comprising
code for presenting a magnified view of a part of an image, in
which a correspondence point is to be defined.
54. The computer program according to claim 49, further comprising
code for presenting the defined transition.
55. The computer program according to claim 54, further comprising
code for halting a presentation of the defined transition at an
arbitrary point.
Description
TECHNICAL FIELD OF THE INVENTION
[0001] The present invention relates to an image-based digital
representation of a scenery, and an image information storing
structure for use in presentation of digital images. The invention
also relates to a method for presentation of a navigation in a
scenery represented by digital images, and a method for
presentation of digital images. The invention further relates to a
device, a method and a computer program for creating an image-based
digital representation of a scenery.
BACKGROUND OF THE INVENTION
[0002] Today, the interest in presenting information on computer
networks, such as the Internet, is quickly increasing. There is a
desire to present information taking up large memory space, such as
images. At the same time, connections to the Internet are now being
made through mobile phones with slower connections and thus longer
downloading times. Therefore, as small data packages as possible
should be downloaded while as much information as possible is
desired to be presented. These two features are very hard to
combine.
[0003] The problem is especially noticeable when images are to be
downloaded, and even more if the image depicts complex objects such
as in a scenery. If a company would like to present a product on
its home page, a simple picture of the product might not be
satisfying. All details of the product might not be evident from a
picture of reasonable storing size. Then, a first picture in low
resolution could be presented and a larger picture in high
resolution could be downloaded only when a user requests it.
However, if the high resolution is to be presented the large size
picture still has to be downloaded with its associated long
downloading time.
[0004] If a landscape is to be presented, the presentation is even
more impeded by a slow connection. A landscape may contain lots of
details, requiring very large images, or even a great number of
images of different viewing angles. Currently, there are methods
for creating a computer representation of 3D sceneries. Then, all
objects in the scenery need to be modelled in the computer. These
methods require large memory capacities for storing the generated
model, and the creation of the computer representation of the
scenery is very complicated and time consuming.
[0005] If several images of different viewing angles are obtained,
these images may be stitched together to create a very large image
for enabling panning in the scenery. Such a method is shown in U.S.
Pat. No. 5,963,664. However, these very large images are not
practical to download through a slow connection. Further, U.S. Pat.
No. 6,198,852 discloses a method for creating a new view of a
scenery from a different viewing angle out of two views of the
scenery from different viewing angles. However, this only provides
a method for creating a new view of the scenery. It will not allow
a user to navigate in the scenery. Further, it will create more
views which would require longer downloading times, since more
images are to be downloaded.
[0006] It may also be convenient to create an animation for showing
a transition between different viewing angles. However, an
animation has an even larger size than a single picture. There will
therefore be no desire to download such an animation through a slow
connection since it will take too much time.
[0007] Some methods are known for generation of image data which is
to be shown in an animation between two images. Such methods are
disclosed e.g. in "Navigating Through Sparse Views" by S.
Fleishman, B. Chen, A. Kaufman, and D. Cohen-Or, VRST (Symposium on
Virtual Reality Software and Technology) '99, and in "Feature-Based
Image Metamorphosis", T. Beier and S. Neely, SIGGRAPH '92.
SUMMARY OF THE INVENTION
[0008] It is an object of the invention to provide improved
presentation of information. The presentation should require short
downloading times and still be able to present a lot of
information. Further, a scenery should be easily represented
digitally for enabling navigation between different views of the
scenery.
[0009] These and other objects are achieved by an image-based
digital representation of a scenery according to claim 1, a method
for presentation of a navigation in a scenery according to claim
22, an image information storing structure according to claims 30
and 31, a method for presentation according to claim 32, and a
device, method and computer program for creating an image
information storing structure according to claims 33, 42, and
49.
[0010] Thus, according to one aspect of the invention, there is
provided an image-based digital representation of a scenery for
visually presenting the scenery. The representation being stored in
an image information storing structure, which comprises vertices,
each representing a digital image, and edges binding together a
first vertex with a second vertex. Each edge represents information
of a transition between a first image in the first vertex and a
second image in the second vertex, wherein the first and the second
image depict different views of the scenery. The first image has a
first area and the second image has a second area, wherein the
depictions of the first area substantially correspond to the
depictions of the second area. The transition information defines a
manipulation of at least one of the first and the second images,
said manipulation comprising a computation of at least one
transition image, the transition image depicting a new view of the
scenery depicted in the first and second images.
[0011] The image-based representation of the scenery implies that a
complex scenery may be easily modelled. Then, through the
manipulation defined by the transition, images of the scenery may
be computed. The storing of the images and information of
transitions in an image information storing structure arranges the
different images in a logical way so that navigation between the
images through transition images may easily be performed.
Furthermore, only the images and the transition information need to
be loaded, e.g. downloaded, in order to view such a navigation,
since they hold all information needed. This implies that a lot of
information may be presented while requiring only small data
amounts to be downloaded. Transition images may be computed at a
presentation device using the defined manipulation, which has been
received through the download. Thus, the presentation may be
performed over slow connections to the image-based representation
without requiring extensive downloading of information over the
connection. Further, since the transition images may be computed at
the presentation device, the presentation device may control what
images are to be shown. This enables free navigation between the
first and the second image, i.e. if the first image is presented
depicting a first view of the scenery and a transition is activated
towards a second view of the scenery depicted in the second image,
the transition may be halted or reversed at any time.
[0012] Furthermore, if the image information storing structure
comprises several vertices and thus several images with connecting
edges a dynamic presentation is enabled. Several transitions back
and forth between different images could be performed, and a
sequence in which the transitions are to be made is not stipulated.
Thus, there is provided a possibility to vary the presentation and
in which order images should be presented.
[0013] It is an advantage of the invention that images, which have
not been stored and need not be downloaded, may be presented. This
enables a presentation of a scene by means of a limited number of
images. Navigation in the scene is still enabled, i.e. an observer
could be guided from a presentation of a first view to a
presentation of a second view by transition images being shown.
Thereby, the observer will get an impression of how the first and
the second views are interrelated. Further, the observer will not
experience any annoying waiting times between the presentations of
the first and the second images. Further, a transition from a first
image to a second image may be halted. A user may then reverse the
direction of the transition back to show the first image or start a
transition in another direction towards a third image. Thus, the
user has large possibilities to control what image is
presented.
[0014] In a specific application of the invention, the second image
could be completely comprised in the first image, and the
transition between the first image and the second image would then
constitute zooming in the first image. In this particular
application, the manipulation could be performed using only the
first image. Thus, only the first image needs to be downloaded in
order to enable the transition. Then, during the transition, the
second image could be downloaded, and the downloading time will not
be noticed.
[0015] In the context of this application, the term "scenery"
should be interpreted broadly as a 3D-environment, including not
only a landscape or a room, but also a 3D-object. Further, the
"scenery" may be a real, existing environment or a virtual, drawn
or computer generated environment. Within the art other common
denominations used to describe such a scenery are scene or
3D-world. The term "view" should be interpreted as the appearance
of the whole or a part of the scenery from a certain position
(viewing point) and angle (viewing angle). The viewing angle
defines an angle between an optical axis of the depiction plane and
the coordinate axes of the 3D-scenery. In the context of this
application, a first image showing a part of a second image in
greater scale is regarded as depicting a different view than the
second image. In this case, the first and the second image have
different viewing points but the same viewing angle.
[0016] A transition image will only depict parts of the scenery
that are depicted in any of the first and second images. The
transition image may depict the scenery from a different viewing
point and/or viewing angle than the viewing points and viewing
angles of the first and second images. Since the transition image
is created through manipulation of at least one other image, the
transition image will not present a perfect depiction of the
scenery, but the depiction of the transition image may be so good
that no faults or only small faults in the transition image are
visible to the human eye. A transition implies a change of viewing
point and/or viewing angle of the presented image in accordance
with the transition information. The transition information
describes how the presented image should be altered for changing
the viewing point and/or viewing angle.
[0017] In the context of this application, the term "area" of an
image should be interpreted as a portion or section of the
two-dimensional image surface depicting a 3D-scenery, not a region
of an object or scenery that is depicted in the image. However, the
portion or section of the image surface forming the "area" will of
course hold information of part of the depictions in the image.
[0018] Depictions of two areas "substantially corresponding" should
be interpreted as the two areas depicting or showing the same or
mainly the same part of the same or at least almost identical
objects or sceneries. For example, a first area depicting a front
view of a painting substantially corresponds to a second area
depicting the painting from a side angle. The second area will show
more of the frame of the painting, but the information in the first
and the second area will still be corresponding.
[0019] Images depicting the view may be obtained in several
different ways, such as by taking photographs with a camera,
drawing views of the scenery, computer generating views of the
scenery, etc. The images may also be obtained by video cameras,
which continuously monitor a landscape or a room. Then, the images
in the image information storing structure will be continuously
replaced or updated. This will enable real time navigation in a
scenery that is monitored by video cameras.
[0020] Further, the image information storing structure describes
the structure of how the images and transitions are interrelated,
it should not be interpreted as a description of how the structure
is implemented.
[0021] According to another aspect of the present invention, there
is provided an image information storing structure for altering a
presentation of a first image so that a transition is made to a
presentation of a second image. The image information storing
structure is accessible by a presentation device and comprises
vertices, each representing an image, and edges binding together a
first vertex with a second vertex. Each edge represents information
of a transition from a first image in the first vertex to a second
image in the second vertex, wherein the first image has a first
area and the second image has a second area, information in the
first area substantially corresponding to information in the second
area. The transition information defines a manipulation of at least
one of the first and the second images, wherein the manipulation
forms a function for altering a presentation of the first image so
that a transfer is made to a presentation of the second image via a
presentation of at least one transition image.
[0022] According to yet another aspect of the invention, there is
provided an image information storing structure, which is
accessible to a presentation device and comprises a first image and
a second image. The first image has a first area and the second
image has a second area, wherein information in the first area
substantially corresponds to information in the second area. The
image information storing structure further comprises information
of a transition from a first image to a second image. The
transition information defines a manipulation of at least one of
the first and the second images, the manipulation forming a
function for altering a presentation of the first image so that a
transfer is made to a presentation of the second image via a
presentation of at least one transition image.
[0023] The image information storing structure according to the
invention enables a transition from a first image to a second image
to be calculated when the transition is to be made. Thus, there is
no need of downloading a file containing a recorded animation, but
instead only a start image and an end image need to be downloaded
together with a definition on how a manipulation is to be made.
This implies that a small amount of data could be downloaded for
presentation of a lot of information. Part of the presented
information could then be created during the presentation by means
of the defined manipulation.
[0024] The needed information is conveniently stored in the image
information storing structure and accessible to a presentation
device. The manipulation could then be used for calculating and
creating a transition from the start image to the end image in the
presentation device or in a computer connected to it. Thus, the
transition is calculated when it is to be performed and only
information that is absolutely needed is downloaded. Information
created by the transition is not downloaded, which shortens the
downloading times.
[0025] According to yet another aspect of the invention, there is
provided a method for presentation of a navigation in a scenery
represented by digital images. The method comprises the steps of
accessing a first and a second digital image, the first and the
second image depicting different views of the scenery. The first
image has a first area and the second image has a second area,
wherein the depictions of the first area substantially correspond
to the depictions of the second area. The method further comprises
the steps of accessing information of a transition between the
first image and the second image, said transition information
defining a manipulation of at least one of the first and the second
images, and determining at least one transition image using the
defined manipulation. The transition image depicts a new view of
the scenery depicted in the first and second images. The method
further comprises the step of performing a visual navigation in the
scenery by replacing a presentation of an image depicting a view of
the scenery by a presentation of another image depicting a
different view of the scenery.
[0026] This implies that an observer may be guided through a
scenery. Different parts of the scenery may be shown and the
observer will get a sense of how the parts of the scenery are
interrelated. Further, this navigation will be achieved by
downloading only small amounts of data, which enables complex
presentations over a slow connection. It is not necessary that the
first and the second images are presented during the navigation. It
may be sufficient to present a navigation in transition images
only. Then the first and second images are used only for the
computation of transition images. Further, if only the first image
is used for the manipulation, it is not necessary to access the
second image until the second image is to be presented.
[0027] According to yet another aspect of the invention, there is
provided a method for presentation of digital images. The method
comprises the step of accessing a first and a second stored digital
image. The first image has a first area and the second image has a
second area, wherein information in the first area substantially
corresponds to information in the second region. The method further
comprises the steps of accessing information of a defined
transition from the first to the second image, presenting the first
image on a presentation device, performing the transition from the
first image to the second image by calculating a manipulation of at
least one of the first and the second images, and presenting at
least one transition image on the presentation device. The
transition image is created by the manipulation. The method further
comprises the step of presenting the second image on the
presentation device to complete the transition of presentation of
the first image to presentation of the second image.
[0028] The method according to the invention allows a transition to
be made from a first image to a second image on a presentation
device without demanding extensive downloading of information.
[0029] According to still another aspect of the invention, there is
provided a device for creating an image-based digital
representation of a scenery for virtually presenting the scenery.
The device comprises means for obtaining a first and a second
digital image, the first and the second image depicting different
views of the scenery, and means for defining a first area of the
first image and a second area of the second image, wherein the
depictions of the first area substantially correspond to the
depictions of the second area. The device further comprises means
for defining information of a transition between the first image
and the second image, wherein the transition information defines a
manipulation of at least one of the first and the second images.
The manipulation comprises a computation of at least one transition
image, the transition image depicting a new view of the scenery
depicted in the first and second images. The device further
comprises means for storing the first and the second images as two
vertices in a graph of an image information storing structure and
the defined transition information from the first to the second
image as an edge binding together said two vertices in the graph of
the image information storing structure.
[0030] This device provides a possibility to easily create a new
image information storing structure, which could be used for the
new method of transforming a presentation of a first image to a
presentation of a second image.
[0031] Preferred embodiments according to the different aspects of
the invention will now be described. It is possible to combine
embodiments presented for one aspect of the invention with other
aspects of the invention.
[0032] According to the image-based digital representation of the
invention, the manipulation preferably comprises a computation of
several transition images depicting different views of the scenery.
This gives larger freedom for navigation in the scenery, since
there are more images available for presentation.
[0033] In a preferred embodiment of the image-based digital
representation of the invention, the transition defines a
manipulation for altering a presentation of a first image, the
depictions of which comprising a planar surface from a first
viewing point, to a presentation of a second image, the depictions
of which comprising said planar surface from a second viewing
point.
[0034] Since the first image and the second image both depict the
same planar surface, a navigation between different viewing points
and viewing angles is easily enabled. An observer will then be
guided by transition images through a change of viewing angle of
the scenery to show the scenery from the other angle.
[0035] Thus, there is no need to download a recorded animation of
transforming a presentation of the scenery from one position to a
presentation of the scenery from another position. Only the start
and end images together with the transition to be performed is
needed. Further, the navigation may be halted at any transition
image. Then, the user may freely choose to continue to perform a
transition to a presentation of the second image or in stead to
initiate a transition towards the presentation of the first image
or towards a third image depicting the scenery from yet another
viewing point or viewing angle.
[0036] In another preferred embodiment, the transition defines a
manipulation for altering a presentation of a first image depicting
a scenery from a viewing point in a first viewing angle to a
presentation of a second image depicting the scenery from said
viewing point in a second viewing angle.
[0037] This would correspond to a rotation of a camera depicting
the scenery. This may preferably be combined with a transition
between different viewing points for allowing even more freedom of
changing the presentation of the scenery.
[0038] In yet another preferred embodiment, the transition defines
a manipulation for altering a presentation of a first image
depicting a scenery from a first viewing point in a viewing angle
to a presentation of a second image depicting the scenery from a
second viewing point in said viewing angle.
[0039] Thus, if a first image representing a product is presented,
details on the product may be presented by performing a transition
towards a presentation of a second image in higher resolution, i.e.
depicting the scenery from a closer viewing point with the same
viewing angle. Thus, the second image represents only the detail.
When the presentation of the first image is transformed into the
presentation of the second image according to the invention, a user
is guided by transition images to the detail and will get an
impression of where the detail is situated in the product. Thus, if
a detail is to be shown, the first image does not have to be
replaced by a large size image in high resolution, which would
require a long downloading time, but instead only a small size
image of the particular detail in high resolution may be presented.
This is a great advantage by comparison with other representations
of a scenery, such as a generated 3D computer model of the scenery.
If it should be possible to present several details, several
separate images each representing one or a few details may be used.
Thus, information that is not of interest will not be downloaded or
stored in high resolution, since no high resolution image will be
available depicting uninteresting details.
[0040] Preferably, the manipulation is based on the information in
the first and second areas. Thus, the manipulation will use
information that is common to the two images. Thereby, the
interrelationship between the images could be determined and
manipulations, such as translations or rotations, needed for
converting the first image to the second image is known.
[0041] Further, the manipulation preferably uses information of
correspondence points in the first and the second areas. Use of
four correspondence points is sufficient for relating the images to
each other according to a projective transformation, which will
transform the image plane of the first image to the image plane of
the second image. This transformation may also be used for
transforming the image planes of the first and second images to an
image plane of a transition image for computing the transition
image.
[0042] The manipulation may comprise a projective transformation.
By means of a projective transformation the first image is
transformed so that the four points of the first image are placed
in a position, which could be chosen arbitrarily. The position is
determined by a function, which calculates intermediate points
between the four correspondence points of the two images. The
projective transformation will alter the conceived viewing point
and/or viewing angle of the first image and during the transition
the viewing point and/or viewing angle will be continuously altered
from the viewing point and viewing angle of the first image to the
viewing point and viewing angle of the second image. Further, using
the projective transformation, the transition may be calculated
very quickly. However, this limits the transitions which are
possible to perform to transitions between images depicting the
same plane, images depicting a scenery from the same point of view
but in different viewing angles, and images having different zoom
levels, which could be approximated as depicting the scenery from
different viewing points but in the same viewing angle. Further,
transitions constituting combinations of the previously mentioned
cases may still be performed.
[0043] Preferably, the transition information defines a function of
the progress of the transition for the movement of the
correspondence points from a position in the first image to another
position in the second image. The movement of the correspondence
points is dependent on the progress of the transition, i.e. how
much the viewing point and viewing angle have changed. The movement
of the correspondence points may then be used for computing where
all other points of the images should be placed in the transition
images.
[0044] In a preferred embodiment, the transition further comprises
a point set, which define the positions of the correspondence
points in a specific transition image. Thus, the transition may be
forced through presenting a transition image with a specific
viewing point and viewing angle. The definition of the point set
may be achieved by using an additional image, depicting a view of
the scene, said view being different from the views depicted in the
first and the second images. The point set may then be defined as
the positions of the correspondence points in the additional image.
The transition between the first and the second image may thus be
forced through the view of the additional image. However, the
additional image need not be stored in the image information
storing structure. The view depicted in the additional image may be
computed through the manipulation of the first and/or the second
image.
[0045] In another preferred embodiment, a point in the transition
image is computable by blending a mapping of a point in the first
image onto said point in the transition image with a mapping of a
point in the second image onto said point in the transition image.
The blending ensures that information of both images may be used.
The blending may be calculated pixelwise by computing the influence
from the first and second images on one point of the transition
image at a time. As an alternative, all points of the first and
second images may be mapped onto the image plane of the transition
image before blending the mapped images.
[0046] The transition information preferably defines weights for
the blending. These weights define the influence of each point in
the first and second images for calculation of the transition
image.
[0047] The transition information preferably defines different
weights for the intermediate images. These weights define the
influence of the intermediate images for calculation of the
transition image. Thus, the influence on the transition image from
the first image could differ from the influence from the second
image. If a transition image in the beginning of the transition is
calculated, the intermediate image calculated from the first image
could have greater influence than the intermediate image calculated
from the second image. This might be advantageous since this
transition image will be most resembling or closest to the first
image. The influence or weights for the intermediate images could
be varied throughout the transition to take into consideration that
the transition images in the beginning of the transition will be
most resembling to the first image and in the end of the transition
will be most resembling to the second image. Preferably, the
weights of the first and the second images vary as a function of
the progress of the transition for calculation of different
transition images.
[0048] If e.g. the brightness differ substantially between the
first and the second images, edge effects may occur in the
transition image. This means that the mapping of an edge of one of
the first and second images onto the transition image will appear
in the transition image as an edge between a darker and a brighter
region in the image. Preferably, the weights differ within the
first and second image in accordance with a predefined function. By
e.g. smoothly decreasing the weight of the points towards the edges
of the first and second image, the mapping of the edges of the
first and the second images will not appear sharply in the
transition image.
[0049] The manipulation may also preferably use at least one
additional image for computation of the transition image, the
first, the second and the additional images depicting different
views of the scenery, the first image having a first area, the
second image having a second area, and the additional image having
a third area, wherein the depictions of the first area
substantially correspond to the depictions of the second area and
the third area. This will enable even more freedom of navigation in
a scenery, since the presented view may be altered between at least
three different views simultaneously. It may also improve the
quality of the presented transition images, since more information
is present in the images used for computing the transition
images.
[0050] Preferably, the transition defines a function for altering a
presentation of at least a part of the first image to a
presentation of at least a part of the second image via a
presentation of the at least one transition image. Then, an
animation may be shown for continuously changing the viewing point
and viewing angle. When only a part of the first and the second
image is presented, the presentation of areas in the transition
image receiving no information from either the first image or the
second image may be avoided. If no point of the projections of the
first image and the second image is projected onto a point of the
transition image, this would be the case.
[0051] Advantageously, the transition is reversible, whereby the
transition is usable for altering a presentation of at least a part
of the second image to a presentation of at least a part of the
first image. Thus, the same transition may be used for the reversed
transition, which implies that only one transition between the
images needs to be defined.
[0052] The vertex representing the first image may further comprise
a link to another vertex via at least one specified edge for
transition of a presentation of the first image to a presentation
of the image represented by the other vertex. As a result, a
navigation through several images depicting different views of the
scenery may be predefined. Upon activation of the link, the
navigation will be presented.
[0053] The link may be coupled to a hotspot or other means of
navigation, such as mouse movement. Then, a user may activate the
navigation by clicking in the hotspot or by moving the mouse. This
could illustrate in what direction the viewing point and viewing
angle will be changed during the navigation. However, the link may
also be coupled to any graphical object presented to the user. For
example, the link may be coupled to a button or a text.
[0054] Preferably, the first and second areas are predefined. Then,
the image information storing structure holds information of which
areas of the images that should be used for the manipulation before
the manipulation is actually performed.
[0055] The transition is preferably initiatable by input from a
user. The user may initiate the transition in a lot of different
ways, such as by pushing a button with a pointing means, e.g. a
mouse, and/or moving the pointing means.
[0056] According to a preferred embodiment of the method of
presentation of a navigation, a user controls what image is to be
presented. This implies that a user will freely control how the
viewing point and viewing angle will be altered when changing the
image which is being presented.
[0057] According to another embodiment, an animation is presented,
said animation forming a specified path for altering the
presentation of an image to the presentation of another image. Then
a specified succession of images will be shown. The path of the
animation defines which transitions that should be used, i.e.
showing a transition from a first image to a second image and then,
possibly, further transitions to additional images.
[0058] According to a preferred embodiment of the method of the
invention, the method further comprises the step of downloading the
first and the second images to the presentation device.
[0059] The downloading of the first image will be performed before
the transition is started. If the manipulation only manipulates the
first image in creating the transition images, the second image
need not be downloaded before the transition is started. In stead,
the second image may be downloaded during the transition and
manipulation of the first image. In this way, the downloading time
of the second image will not be noticed, since a transition from a
presentation of the first image is presented during the download.
This is particularly useful when the transition constitutes a
zooming in the first image, since then the second image is
completely incorporated in the first image. This may also be used
for a transition where the viewing angle and the viewing position
are changed, if all of the depiction of the second image is within
the depiction of the first image.
[0060] The method may further comprise the step of enabling a
transition from a first image to a second image, when both images
have been downloaded to the presentation device. Thus, a transition
may be activated when the images needed for performing the
manipulation have been downloaded. The calculations of the
manipulation is preferably performed at the presentation device.
Then, data created in the manipulation need not be downloaded
through a slow connection, but is created where it is needed.
[0061] The steps of accessing the first and the second images and
the transition information comprise the step of accessing a
computer program, which reads a parameter file comprising a
description of the image information storing structure. The
computer program will then be able to find the needed or desired
images that are to be accessed or downloaded from the storing
structure or from a location pointed out by the structure. The
computer program preferably reads a description of the image
information storing structure. This description could for example
be in XML (eXtensive Markup Language). However, a description in
any other suitable language may be read.
[0062] According to a preferred embodiment of the device for
creating an image information storing structure of the invention,
the device further comprises means for presenting an overview of
the graph of the image information storing structure. This enables
an operator who creates an image information storing structure to
easily embrace the structure that has been created.
[0063] Preferably, the device further comprises means for defining
correspondence points of the first and the second images for use in
the manipulation. Then, suitable points that will give a transition
with good quality of the transition images may be defined.
[0064] The device may also comprise means for presenting guiding
lines for the definition of corresponding points. These guiding
lines may help the user to correctly define the points and to
better see how the points are related to each other.
[0065] Preferably, the device further comprises means for
presenting a magnified view of a part of an image, in which a
corresponding point is to be defined. This makes it even easier for
the operator to correctly define the correspondence points.
Preferably, the device further comprises means for scaling down the
obtained images in resolution. This enables the images that are to
be stored into the image structure to be packed to a small format.
Also the means for presenting a magnified view may then use the
obtained image, which is in high resolution.
[0066] The device may further comprise means for presenting the
defined transition. This enables the operator to instantly check
the quality of the defined transition.
[0067] Preferably, the device further comprises means for halting a
presentation of the defined transition at an arbitrary point. Then,
the operator can check the quality of each part of the transition.
Since the transition is freezed, the operator may study each part
of the transition thoroughly.
BRIEF DESCRIPTION OF THE DRAWINGS
[0068] The invention will now be described in more detail by way of
example referring to the appended drawings.
[0069] FIG. 1 is a schematic view of an image information storing
structure according to the invention and its accessibility.
[0070] FIG. 2 is a view over images of an image information storing
structure.
[0071] FIG. 3 is a schematic view of a mapping of a plane in a
first image to a plane in a second image.
[0072] FIG. 4 is a schematic illustration of a path of
transition.
[0073] FIG. 5 is a schematic view of a blending between a mapping
of the first image and a mapping of the second image.
[0074] FIG. 6 is a flow chart illustrating a method of presenting
images according to the invention.
[0075] FIG. 7 is a schematic view of areas in an image for
activating transitions.
[0076] FIG. 8 is a schematic view of a first mode of a device for
creating an image information storing structure according to the
invention.
[0077] FIG. 9 is a schematic view of a second mode of a device for
creating an image information storing structure according to the
invention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
[0078] Referring to FIG. 1, access to a presentation of images is
described. An image information storing structure 1 is stored in
such a way that it is accessible for a presentation device 3. Thus,
the image information storing structure may be stored on a server
2, or locally on a storing unit. It may also be embedded in the
presentation device or in a viewer software. The access could be
achieved over a computer network 4, such as the Internet, or over a
file system if the image information storing structure is stored
locally. The computer network 4 could also be wireless, such that
the presentation device 3 could be constituted of a mobile
phone.
[0079] The server 2, which may be accessed through a home page,
comprises the image information storing structure 1. This image
information storing structure 1 may comprise pointers or references
to different images and transitions between these images. Thus, the
actual images and transitions need not be stored on the server 2.
The image information storing structure 1 may alternatively
comprise the actual images and transitions. In the following, the
image information storing structure 1 will be described as
comprising the actual images and transitions.
[0080] Referring to FIG. 2, the image information storing structure
1 according to the invention is described in more detail. The image
information storing structure 1 comprises at least a first 6 and a
second 7 image and information of a defined transition 8 for
transforming a presentation of the first image 6 to a presentation
of the second image 7. However, the image information storing
structure 1 preferably comprises several images and information of
transitions between these images. The construction of the structure
could be described as a graph, i.e. vertices connected by edges.
Each vertex comprises one image, and each edge comprises
information of a transition between the two images, which are
connected by the edge.
[0081] The first 6 and the second images 7 comprise areas 9, 10,
the information of which substantially corresponds. These areas 9,
10 having corresponding information are used for determining a
transformation of a presentation of the first image 6 to a
presentation of the second image 7. The corresponding information
may constitute an object or surface, which is depicted in both
images. It may also constitute depictions of corresponding objects
in the first 6 and second images 7. All or only a part of the
corresponding information is used for the transition. In the
preferred embodiment, only four correspondence points are used for
the transition. The corresponding information preferably comprises
a plane, i.e. a planar surface defined in the depiction of the
images. Through this plane, the relation between the first 6 and
the second images 7 may be determined by means of the plane
defining a projective mapping from the first image 6 to the second
image 7. This relation could then be used for the transition 8. It
is sufficient that the common planar surface constitute only a
small part of the images 6, 7. Transitions 8 may also be defined
between images that depict a scenery from the same viewing point
but in different viewing angles, or for zooming between images,
where one of the images shows a detail of the other image in higher
resolution.
[0082] A transition from the first image 6 to the second image 7
constitutes a defined manipulation of at least one of the images
for smoothly transforming a presentation of the first image 6 to a
presentation of the second image 7. The manipulation being defined
implies that the intended path of the transition is known. However,
the actual calculation will not be performed until the transition
from the presentation of the first image 6 to the presentation of
the second image 7 is actually made. The manipulation will use
corresponding information of the images. The corresponding
information could be four correspondence points defining a plane
that is depicted in the images.
[0083] Links to other images may be coupled to each image. These
links constitute pointers to a path through the graph, following
specific edges for transforming the presentation. Thus, the link
could activate a transition 8 from the first image 6 to the second
image 7 via the transition 8. However, it could also activate a
transition from the first image to the third image via a first
transition to the second image and a second transition to the third
image. The link specifies a certain path to be taken between the
images, i.e. what edges should be followed. The links could be
coupled to a hotspot or other means of navigation, such as mouse
movement. Thus, a user could activate the transition by activating
the area to which the link is coupled. Alternatively, the link
could be coupled to a specific part of the image, such as an arrow
11 in the image or an area of the image, which may be depicting the
same parts of the scenery as the image to which the transition is
made. A link could also be created by requesting a graph search for
a certain vertex from another vertex, for example from the vertex
that is presented at the time of the search.
[0084] Referring to FIG. 3, a manipulation will be described. A
projected plane 12 is determined in a first 6 and a second image 7.
The plane 12 is defined by four points 13-16. These points 13-16
are corresponding in the two images 6, 7. The correspondence points
13-16 are used to compute a homography. Homographies are described
in more detail in Paul Heckbert; "Projective Mappings for Image
Warping"; Rendering, 15-862; Sep. 26, 1995. The correspondence
points 13-16 are named u.sub.i, v.sub.i in the first image and
x.sub.i, y.sub.i in the second image for i=0,1,2,3. A homography
matrix H is computed by solving the following 8.times.8 equation
system. 1 [ u i v i 1 0 0 0 - u i x i - v i x i 0 0 0 u i v i 1 - u
i y i - v i y i ] [ a b c d e f g h ] = [ x i y i ] H = [ a b c d e
f g h l ]
[0085] The homography matrix is used to transform points in the
first image 6 to points in the second image 7. In a preferred
embodiment, the transform is passed through a [0,0,1,1] plane 12.
This gives simpler calculations. In FIG. 3, the mapping of a plane
12 in the first image 6 onto the [0,0,1,1] plane 12 and then onto a
plane 12 in the second image 7 is illustrated. There are several
other methods for transforming the images. The transform may be
based on information other than the correspondence points, such as
the image contents in the areas of the images comprising
corresponding information.
[0086] In order to create a smooth transform from the first image 6
to the second image 7, an interpolation is performed in conjunction
with the projective transformation. Each correspondence point 13-16
follows a path from the position in the first image 6 to the
position in the second image 7. The path may be specified using key
point sets. The key point sets define positions which the
correspondence points 13-16 should have for a specific transition
image. Thus, the first image 6 and the second image 7 are the first
and the last key point sets in the transform. More key point sets
can be inserted between the first and last key point sets to
control the path of the correspondence points 13-16. Certain
positions of an interpolation of the correspondence points 13-16
between the key point sets are used to calculate transition images,
which are presented during the transform. These transition images
are presented in succession to create the smooth transform.
[0087] The transform preferably describes a circular or arcuate
path for the correspondence points 13-16, when a transition
corresponding to a rotation of a camera is performed. This is
advantageous, since an object which is rotated between the two
images will not appear to be distorted during the transform. If a
linear path would be used, the object will be smaller during the
transform. Thus, this will not result in a rigid transform.
Further, the key point sets may be defined in such manner that the
transition will not create transition images, which require
information that is not present in any of the first and the second
images. This may otherwise be at risk, if a linear interpolation is
made between the first 6 and the second images 7, as shown in FIG.
4, indicated by arrows A. If a key point set is used instead, the
alternative path indicated by arrows B may be defined, and the
transition images will not include any information which is not
present in the first 6 and the second images 7. In other cases of
transitions, it may be suitable to use a linear path.
[0088] Further, if only a part of the first 6 and second images 7
are shown during the transition the risk of showing information
which is not present in any of the first 6 and the second images 7
may be avoided. This would correspond to a zooming in the first 6
and the second images 7. Then, the mappings of the first 6 and the
second images 7 hold more information than presented when the first
6 and the second images 7 are presented. As a result, it may be
assured that at least one pixel of the first 6 and the second
images 7 is mapped onto each pixel of each transition image.
[0089] The position of the parameters, i.e. the correspondence
points 13-16, in a transition image is a function of the progress
of the transition. The positions may be calculated by means of an
interpolation equation, for example a spline equation.
[0090] The interpolation process results in three point sets, i.e.
the point set of the first image 6, the point set of the second
image 7, and the point set of the transition image to be generated.
Using these point sets two homographies can be computed, one from
the first image 6 to the transition image, and one from the second
image 7 to the transition image. By applying these homographies to
the first 6 and the second images 7, respectively, each pixel of
the first 6 and second images 7 may be projected onto the image
plane of the transition image to be generated. These projections
are combined to a final transition image by blending the
projections from the first and the second images in each pixel of
the transition image.
[0091] Referring to FIG. 5, a blending of the first 6 and second
images 7 into a transition image is described. Each pixel, i.e. a
discrete point in a two-dimensional coordinate system, of the first
6 and the second images 7 is transformed by multiplication of the
corresponding homography matrix. Then, a combination or blending of
the projections of the first 6 and the second images 7 is used for
calculating the transition image. Each pixel of the original images
(the untransformed first and second images) is assigned a weight
value. In a preferred embodiment of the invention, a linear
combination of the pixels from the first 6 and second images 7 is
used for determining the values of pixels in the transition image.
Then, the blending could be described as: 2 P = + P 1 + + P 2 ,
[0092] where P is the value of the pixel in the transition image,
P.sub.1 is the value of the pixel in the first image 6, which is
mapped onto the pixel P in the transition image, and P.sub.2 is the
value of the pixel in the second image 7, which is mapped onto the
pixel P in the transition image. The weights of the pixels from the
first 6 and the second images 7 are determined by the parameters
.alpha. and .beta.. In a preferred embodiment, these weights differ
depending on the progress of the transition. Thus, the parameter
.alpha. should be large if the transition is close to the first
image 6 and the parameter .beta. should be large if the transition
is close to the second image 7. This could be described as:
.alpha.=f(u,v,1-t),
.beta.=f(u,v,t),
[0093] where t=[0,1] is a parameter describing the progress of the
transition. As seen, the parameters .alpha. and .beta. also depend
on a function f. This depends on the position of the pixel in the
first and the second images. Thus, the weight of a pixel is
dependent on where it is situated in the image. Preferably, the
function f is a two-dimensional bell-curve, which would give large
weights to pixels in the middle of an image and small weights to
pixels in the edge of an image. Thereby, edge effects as described
earlier may be avoided. Thus, the function f could be described
as:
f(u,v)=1-(x.sup.2(3-2x)).sup.2,
x={square root}{square root over
((u-0,5).sup.2+(v-0,5).sup.2)},
[0094] where u, v=[0,1] is the coordinates of the pixel in the
image to be transformed.
[0095] The correspondence points 13-16 need not be used for the
whole transition from a first image 6 to a second image 7. Instead,
at a certain stage of the transition, another set of correspondence
points may replace the first set used. Alternatively, more than one
set of correspondence points 13-16 may be used for the calculation
of each transition image. Then, the different sets of
correspondence points are used for computing intermediate
transition images according to the above, and the final transition
image is computed by blending the intermediate transition
images.
[0096] The images may relate to each other as showing a scenery in
different viewing angles from the same point of view. The
transition will then constitute a rotation of the viewing angle.
The images may also be related by showing the same plane from
different viewing points. This plane may be a small planar part of
a curved surface. The different viewing points may show the plane
from different angles or from different distances. The images may
also be related by having different focuses on a scenery showed
from the same point of view and the same viewing angle, for example
the first image may depict the scene focussing on an object close
to the viewing point whereas the second image depicts the scene
focussing on an object farther away from the viewing point. For all
these alternatives, both images are needed for the manipulation.
The images may also be related by the second image showing only a
part of the first image, but in greater resolution. In this case,
the transition may be performed by only manipulating the first
image. Then, a zooming is made in the first image towards the part
of the first image shown in the second image. In this case, the
transition may be started even before both images have been
downloaded. The above described transitions may be combined for
transforming between images related in other ways.
[0097] Referring to FIG. 6, a method for presenting a transition
between images is described. The presentation is performed on a
presentation device. The presentation device accesses an image
information storing structure through a computer network, step 20,
or through a local file structure. The images and transitions of
the image information storing structure are downloaded to the
presentation device, step 22. A part or the whole of a first image
is then presented on the presentation device, step 24. When a
second image and a transition from the first image to the second
image has been downloaded, the transition is enabled. If a user
requests a transition, step 26, the manipulation is calculated,
step 28, at the presentation device by means of the downloaded
transition. The transition presents transition images, step 30,
which guide the user from the first image to the second image. When
the transition is complete, a part or the whole second image is
presented, step 32. When the second image is presented, a new
transition may be initiated to a third image or back to the first
image. Then, transitions may be made in arbitrary order between the
images of the image information storing structure. Also, a
transition may be halted when presenting a transition image. Then,
the transition may reversed towards presenting the first image
again, continued towards presenting the second image, or a new
transition may be started towards presenting a third image.
[0098] Referring to FIG. 7, an image 40 could be divided into
several areas 41-43, which will lead to the transition to different
images. When a pointer is moved into an area the transition is
started. Preferably, the user will also have to click the pointer
to start the transition. However, the transition may be stopped by
moving the pointer out of the area. Then, the initiated transition
is reversed from a presentation of the last presented transition
image to a presentation of the origin image 40. In a preferred
embodiment, the transition may only be stopped up to a certain
point. Then, if the transition has reached too far away from the
origin image 40 it will be completed irrespective of movements of
the pointer. This point may be specified by the parameter t. When
the transition is within the limits of still being able to be
reversed, the transition may also be altered to a transition from
the origin image 40 to another image. As shown in FIG. 7, the
origin image 40 is divided into three areas 41-43 leading to three
different transitions from the origin image 40. A transition to
move to the left from the origin image 40, as indicated by arrow C,
may then be altered into a transition upwards, as indicated by
arrow D, when the parameter t of the transition to the left is
below a critical value.
[0099] If several images are overlapping each other, an almost free
movement in the images is enabled. This is possible if the
transitions are held below the critical value of t.
[0100] Referring to FIGS. 8 and 9, an editor for creating an image
information storing structure will now be described. However,
before the image information storing structure may be created,
images having areas with corresponding information need to be
captured in digital form. These images may then be used to create
the image information storing structure.
[0101] The editor comprises two modes. In a first mode, shown in
FIG. 8, the images may be schematically related to each other.
Here, the images which are to be used in the image information
storing structure are logically related to each other, i.e. the
connections between the images are shown. In this mode, an operator
creating the image information storing structure may get an
overview of the structure. The images to be used are shown as
thumbnail images 50 and connected by lines 51 illustrating the
possible transitions.
[0102] By activating an image or a transition, e.g. by clicking
with a mouse pointer on any of these, the operator is transferred
to a second mode of the editor, as shown in FIG. 9. In this mode,
the operator may study the transition and create new or improve
existing transitions. The presented information in the second mode
is divided in four parts 52-55. In the two upper parts 52, 53, the
two images involved in the transition of current interest are
presented. The operator may define correspondence points 13-16 by
clicking in these two images. To help the operator to correctly
define the correspondence points, guiding lines 56 may be shown in
the images, so that the area defined by the points is
illustrated.
[0103] To further help the operator to correctly define the
correspondence points 13-16, a zoom of the active image is shown in
the right hand lower part 54 of the second mode. This zoom may be
enabled by the image being obtained in higher resolution than the
image that is stored into the image information storing structure.
Then, the high resolution obtained image is scaled down for the
storage of the image into the image information storing structure.
The definition of the correspondence points may also be corrected
in the zoomed-in image.
[0104] Also, an additional image may be used for defining a key
point set, i.e. a set of positions of the correspondence points
13-16 in a certain transition image. The operator may define the
positions of the correspondence points 13-16 as the key point set.
Then, during the transition between a first and a second image, a
transition image will show the view depicted in the additional
image. The additional image need not be stored in the image
information storing structure, but could be used only for the
purpose of defining the key point set.
[0105] Further, in the left hand lower part 55 of the second mode,
the transition resulting from the defined correspondence points may
be viewed. This transition may be viewed as an animation. A free
navigation through the transition is also allowed. As a result, the
operator may calmly examine all parts of the transition to identify
any unsatisfactory parts of the transition.
[0106] The operator may shift between the two modes of the editor
to relate the images to each other in a logical way and to create
all transitions that are needed. When the structure is complete,
the editor will store the images and transitions with the desired
relations between images and transitions maintained.
[0107] It should be emphasized that the preferred embodiments
described herein is in no way limiting and that many alternative
embodiments are possible within the scope of protection defined by
the appended claims. For example, the presentation of the images
may be performed on a presentation device which is directly
connected to the computer on which the image information storing
structure is stored.
[0108] Further, the images may be captured from live cameras which
are filming a scenery from different point of views. Still pictures
captured from the cameras could be used for the navigation in the
scenery. The cameras could then continuously replace the still
picture used for the navigation. Then, the areas of the images
comprising corresponding information may not be predefined.
However, if the scenery has an easily defined plane, such as a wall
or a door or the like, the areas may be defined automatically when
the transition is to be performed. Also, if the same scenery is
being continuously monitored, the same correspondence points may
always be used among objects that are constantly depicted, such as
a door.
[0109] Moreover, other manipulations of the images may be performed
simultaneously with the manipulations for computing the transition
images. Thus, the brightness or coloration of an image may be
manipulated during a transition.
[0110] Also, the order for downloading images of an image
information storing structure may be controlled by keeping
statistics of the probability that an image is viewed by a user.
Thus, if an image information storing structure contains a large
amount of images, the most popular images are downloaded first.
* * * * *