Systems and methods for object detection
A computing system including a processing circuit in communication with a camera having a field of view. The processing circuit is configured to perform operations related to detecting, identifying, and retrieving objects disposed amongst a plurality of objects. The processing circuit may be configured to perform operations related to object recognition template generation, feature generation, hypothesis generation, hypothesis refinement, and hypothesis validation.
1. A computing system configured to generate an object recognition template set for identifying an object in a scene comprising:
at least one processing circuit configured for:
obtaining registration data of the object, the registration data including an object model representative of the object;
determining a plurality of viewpoints of the object model in a three-dimensional space, wherein each of the plurality of viewpoints corresponds to a camera location based on a symmetry of the object;
estimating a plurality of appearances of the object model at each of the plurality of viewpoints;
generating a plurality of object recognition templates according to the plurality of appearances, each of the plurality of object recognition templates corresponding to a respective one of the plurality of appearances, and each of the plurality of object recognition templates corresponding to one viewpoint of the plurality of viewpoints and including an appearance of the object from the one viewpoint; and
communicating the plurality of object recognition templates as the object recognition template set to a robotic control system, wherein the plurality of object recognition templates is for an object picking operation for the object in the scene by a robot interaction, and
wherein each of the plurality of object recognition templates represent a pose the object may have relative to an optical axis of a camera generating image information of the object within the scene.
2. The computing system of claim 1 , wherein each of the plurality of viewpoints further corresponds to a camera rotation angle.
3. The computing system of claim 1 , wherein the object model is fixed within the three-dimensional space.
4. The computing system of claim 1 , wherein the three-dimensional space is substantially spherical and the object model is fixed at a center of the three-dimensional space.
5. The computing system of claim 1 , wherein the plurality of viewpoints are selected according to an even distribution across a virtual surface.
6. The computing system of claim 2 , wherein each camera location corresponds to a set of viewpoints, each viewpoint of the set of viewpoints corresponding to a different camera rotation angle.
7. The computing system of claim 2 , wherein a subset of the plurality of object recognition templates includes object recognition templates corresponding to viewpoints corresponding to different locations and to different camera rotation angles.
8. The computing system of claim 1 , further including:
determining the plurality of viewpoints based on predicted ranges of poses observed for the plurality of object recognition templates.
9. The computing system of claim 1 , further including determining symmetry of the object according to at least one of a determination that an object appearance of the object changes after rotation and an identification of an axis of the object.
10. A method of generating an object recognition template set for identifying an object in a scene comprising:
obtaining a registration data of the object, the registration data including an object model representative of the object;
determining a plurality of viewpoints of the object model in a three-dimensional space, wherein each of the plurality of viewpoints corresponds to a camera location based on a symmetry of the object;
estimating a plurality of appearances of the object model at each of the plurality of viewpoints;
generating a plurality of object recognition templates according to the plurality of appearances, each of the plurality of object recognition templates corresponding to a respective one of the plurality of appearances, and each of the plurality of object recognition templates corresponding to one viewpoint of the plurality of viewpoints and including an appearance of the object from one viewpoint; and
communicating the plurality of object recognition templates as the object recognition template set to a robotic control system, wherein the plurality of object recognition templates is for an object picking operation for the object in the scene by a robot interaction,
wherein each of the plurality of object recognition templates represents a pose the object may have relative to an optical axis of a camera generating image information of the object within the scene.
11. The method of claim 10 , further including:
corresponding each of the plurality of viewpoints to a camera rotation angle.
12. The method of claim 10 , further including:
fixing the object model within the three-dimensional space.
13. The method of claim 10 , further including:
selecting the plurality of viewpoints according to an even distribution across a virtual surface.
14. The method of claim 10 , further including:
determining the plurality of viewpoints based on predicted ranges of poses observed for the plurality of object recognition templates.
15. The method of claim 10 , further including:
determining symmetry of the object according to at least one of a determination that an object appearance of the object changes after rotation and an identification of an axis of the object.
16. A non-transitory computer readable medium, configured with executable instructions for implementing a method for generating an object recognition template for identifying an object in a scene, operable by at least one processing circuit via a communication interface configured to communicate with a robotic system, the method comprising:
receiving a registration data of the object, the registration data including an object model representative of the object;
performing an operation to generate a plurality of viewpoints of the object model in a three-dimensional space, wherein each of the plurality of viewpoints corresponds to a camera location based on a symmetry of the object;
performing an operation to estimate a plurality of appearances of the object model at each of the plurality of viewpoints;
performing an operation to generate a plurality of object recognition templates according to the plurality of appearances, each of the plurality of object recognition templates corresponding to a respective one of the plurality of appearances, and each of the plurality of object recognition templates corresponding to one viewpoint of the plurality of viewpoints and including an appearance of the object from the one viewpoint; and
outputting the plurality of object recognition templates as the object recognition template set to the robotic system, wherein the plurality of object recognition templates is for an object picking operation for the object in the scene by a robot interaction;
wherein each of the plurality of object recognition templates represent a pose the object may have relative to an optical axis of a camera generating image information of the object within the scene.
17. The non-transitory computer readable medium of claim 16 , wherein each of the plurality of viewpoints further corresponds to a camera rotation angle.
18. The non-transitory computer readable medium of claim 17 , wherein a subset of the plurality of object recognition templates includes object recognition templates corresponding to viewpoints corresponding to different locations and to different camera rotation angles.
19. The non-transitory computer readable medium of claim 16 , further including:
performing an operation to generate the plurality of viewpoints based on predicted ranges of poses observed for the plurality of object recognition templates.
20. The non-transitory computer readable medium of claim 16 , further including performing an operation to determine the symmetry of the object according to at least one of a determination that an object appearance of the object changes after rotation and an identification of an axis of the object.