Method and system for creating an annotated object model for a new real-world object
A method and corresponding system for creating an annotated object model of a real world object. The method comprises providing an initial object model for an object for which an annotated object model shall be created, predicting properties of the object, visualizing a representation of the object based on the initial object model wherein the predicted object properties are displayed associated with the representation, and obtaining selection information based on at least one of a user gesture, user pointing operation, user speech input, and user gaze perceived by a user perception device. The method determines a portion of the object corresponding to the selection information, receives property information from a user input and associates the input property information with the corresponding portion of the object.
1. A method for creating an annotated object model of a real world object, the method comprising the following method steps:
providing an initial object model for an object for which an annotated object model is to be created,
predicting properties of the object by generating object properties for the initial object model by morphing a template model onto the initial object model and transferring properties known for the template model as predicted properties for the initial object model,
wherein the properties and the predicted properties include at least one of a designation of parts of the object, a segmentation of the object defining different parts of the object, affordances of the object that comprise at least one of graspable, place-able, and removable, relations between object parts of the object, appearance properties of the object including at least one of a color, material properties of the object comprising at least one of wood, steel, and glass, and relations between multiple objects,
visualizing a representation of the object based on the initial object model wherein the predicted properties of the object are displayed associated with the representation,
obtaining selection information based on at least one of a user gesture, user pointing operation, user speech input, and user gaze perceived by a user perception device,
determining a portion of the object corresponding to the selection information,
receiving property information from a user input,
associating the input property information with the corresponding portion of the object.
2. The method according to claim 1 , wherein
the representation is visualized using an augmented reality (AR) display or a virtual reality (VR) display.
3. The method according to claim 1 , wherein
the selection information is enhanced by a command input by an operator.
4. The method according to claim 1 , wherein
the selection information comprises at least first selection information and second selection information, and the property information defines a relation between portions of the object corresponding to the at least first and second selection information or a property common to the portions.
5. The method according to claim 1 , wherein
the determination of the portion of the object depends on at least one of a type of a gesture, a position of collision between the representation of the object and a perceived operator hand, and a position on the displayed representation pointed at.
6. The method according to claim 5 , wherein
in case of a collision between the representation of the object and the perceived operator hand is determined, a feedback is provided to an operator.
7. The method according to claim 1 , wherein
the predicted properties include at least a definition of segments defining parts of the object.
8. The method according to claim 1 , wherein
the predicted properties include at least a definition of segments and the segments of the object are visualized as an overlay over the representation of the object.
9. The method according to claim 1 , wherein
the associated display of the predicted object properties displays each predicted property in spatial relation to a respective portion of the object for which the property of the object is predicted, and the representation of the object together with the associated predicted properties of the object is manipulated according to a manipulation input received from an operator.
10. The method according to claim 1 , wherein
based on a perceived operator input, the initial object model is adapted and the displayed representation is updated accordingly.
11. The method according to claim 1 , wherein
the initial object model is analyzed by an object part detector for automated segmentation of the object based at least on one of database information and previously received information from an operator during creation of an annotated object model process.
12. A system for creating an annotated object model of a real world object, the system includes
a processor,
an output device, and
an operator perception device,
the processor being configured to be provided with an initial object model for an object for which an annotated object model is to be created,
to predict properties of the object by generating object properties for the initial object model by morphing a template model onto the initial object model and transferring properties known for the template model as predicted properties for the initial object model,
wherein the properties and the predicted properties include at least one of a designation of parts of the object, a segmentation of the object defining different parts of the object, affordances of the object that comprise at least one of graspable, placeable, and removable, relations between object parts of the object, appearance properties of the object including at least one of a color, material properties of the object comprising at least one of wood, steel, and glass, and relations between multiple objects,
to control the output device to visualize a representation of the object based on the initial object model, wherein the predicted properties of the object are displayed associated with the representation,
to determine a portion of the object corresponding to selection information obtained by the operator perception device based on at least one of a user gesture, user pointing operation, user speech input, and user gaze perceived by the operator perception device, and
to associate property information input by an operator with the corresponding portion of the object.
13. A system according to claim 12 , wherein the output device comprises an augmented reality (AR) display or a virtual reality (VR) display.
14. A system according to claim 12 , wherein the operator perception device comprises at least a camera for perceiving operator movements.
15. A system according to claim 12 , wherein the operator perception device comprises at least a microphone.
16. A system according to claim 12 , wherein the system comprises a feedback device for informing the operator of a determined contact with the visualized presentation of the object.
17. A system according to claim 12 , wherein the processor is connected to a database and the processor is configured to store the annotated object model in the database.