Aspects of the subject disclosure may include, for example, a method comprising obtaining, by a processing system including a processor, first and second models for a structure of an object, based respectively on ground-level and aerial observations of the object. Model parameters are determined for
Aspects of the subject disclosure may include, for example, a method comprising obtaining, by a processing system including a processor, first and second models for a structure of an object, based respectively on ground-level and aerial observations of the object. Model parameters are determined for a three-dimensional (3D) third model of the object based on the first and second models; the determining comprises a transfer learning procedure. Data representing observations of the object is captured at an airborne unmanned aircraft system (UAS) operating at an altitude between that of the ground-level observations and the aerial observations. The method also comprises dynamically adjusting the third model in accordance with the operating altitude of the UAS; updating the adjusted third model in accordance with the data; and determining a 3D representation of the structure of the object, based on the updated adjusted third model. Other embodiments are disclosed.
대표청구항▼
1. A method comprising: obtaining, by a processing system including a processor, a first model for a structure of an object, wherein the first model is based on a first plurality of observations comprising ground-level or near-ground-level observations of the object;obtaining, by the processing syst
1. A method comprising: obtaining, by a processing system including a processor, a first model for a structure of an object, wherein the first model is based on a first plurality of observations comprising ground-level or near-ground-level observations of the object;obtaining, by the processing system, a second model for the structure of the object, wherein the second model is based on a second plurality of observations comprising aerial observations of the object;determining, by the processing system, model parameters for a three-dimensional (3D) third model of the object based on the first model and the second model, wherein the determining comprises a transfer learning procedure using a manifold;obtaining, by the processing system, data representing a third plurality of observations of the object, wherein the data is captured at an airborne unmanned aircraft system (UAS) operating at an altitude greater than that of the first plurality of observations and less than that of the second plurality of observations;dynamically adjusting, by the processing system, the third model in accordance with the operating altitude of the UAS, resulting in an adjusted third model;updating, by the processing system, the adjusted third model in accordance with the data, resulting in an updated adjusted third model; anddetermining, by the processing system, a 3D representation of the structure of the object, based on the updated adjusted third model. 2. The method of claim 1, wherein the operating altitude of the UAS is in an altitude range from about 100 feet to about 2000 feet. 3. The method of claim 2, further comprising: predicting, by the processing system, a new operating altitude of the UAS; anddetermining, by the processing system, a predicted 3D representation of the structure of the object in accordance with the new operating altitude. 4. The method of claim 1, wherein the manifold is a Grassmannian manifold. 5. The method of claim 1, further comprising: determining, by the processing system, correspondence parameters representing a correspondence between the data and the third model; andupdating, by the processing system, the correspondence parameters in accordance with the data. 6. The method of claim 5, wherein updating the correspondence parameters comprises adjusting, by the processing system, the correspondence parameters to reduce an error between the third model and the data. 7. The method of claim 1, wherein the third model varies non-linearly with the operating altitude of the UAS, and wherein variation of the third model with the operating altitude corresponds to a non-linear path on the manifold. 8. The method of claim 7, wherein the non-linear path has a first endpoint corresponding to the first model and a second endpoint corresponding to the second model. 9. The method of claim 1, wherein the data captured at the UAS comprises two-dimensional (2D) video images of the object. 10. The method of claim 9, wherein the updating comprises comparing the adjusted third model with live video images of the object, the updating accordingly being performed in real time. 11. A device comprising: a processing system including a processor; anda memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations comprising:obtaining a first model for a structure of an object, wherein the first model is based on a first plurality of observations comprising ground-level or near-ground-level observations of the object;obtaining a second model for the structure of the object, wherein the second model is based on a second plurality of observations comprising aerial observations of the object;determining model parameters for a three-dimensional (3D) third model of the object based on the first model and the second model, wherein the determining comprises a transfer learning procedure using a manifold;obtaining data representing a third plurality of observations of the object, wherein the data is captured at an airborne unmanned aircraft system (UAS) operating at an altitude greater than that of the first plurality of observations and less than that of the second plurality of observations, wherein the data comprises two-dimensional (2D) video images of the object;determining correspondence parameters representing a correspondence between the data and the third model;updating the correspondence parameters in accordance with the data;dynamically adjusting the third model in accordance with the operating altitude of the UAS, resulting in an adjusted third model;updating the adjusted third model in accordance with the data, resulting in an updated adjusted third model; anddetermining a 3D representation of the structure of the object, based on the updated adjusted third model. 12. The device of claim 11, wherein the operating altitude of the UAS is in an altitude range from about 100 feet to about 2000 feet. 13. The device of claim 11, wherein the manifold is a Grassmannian manifold. 14. The device of claim 11, wherein updating the correspondence parameters comprises adjusting, by the processing system, the correspondence parameters to reduce an error between the third model and the data. 15. The device of claim 11, wherein the third model varies non-linearly with the operating altitude of the UAS, and wherein variation of the third model with the operating altitude corresponds to a non-linear path on the manifold. 16. A non-transitory machine-readable medium comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations comprising: obtaining a first model for a structure of an object, wherein the first model is based on a first plurality of observations comprising ground-level or near-ground-level observations of the object;obtaining a second model for the structure of the object, wherein the second model is based on a second plurality of observations comprising aerial observations of the object;determining model parameters for a three-dimensional (3D) third model of the object based on the first model and the second model, wherein the determining comprises a transfer learning procedure;obtaining data representing a third plurality of observations of the object, wherein the data is captured at an airborne unmanned aircraft system (UAS) operating at an altitude greater than that of the first plurality of observations and less than that of the second plurality of observations;dynamically adjusting the third model in accordance with the operating altitude of the UAS, resulting in an adjusted third model;updating the adjusted third model in accordance with the data, resulting in an updated adjusted third model; anddetermining a 3D representation of the structure of the object, based on the updated adjusted third model. 17. The non-transitory machine-readable medium of claim 16, wherein the transfer learning procedure is performed using a Grassmannian manifold. 18. The non-transitory machine-readable medium of claim 17, wherein the third model varies non-linearly with the operating altitude of the UAS, and wherein variation of the third model with the operating altitude corresponds to a non-linear path on the manifold. 19. The non-transitory machine-readable medium of claim 18, wherein the non-linear path has a first endpoint corresponding to the first model and a second endpoint corresponding to the second model. 20. The non-transitory machine-readable medium of claim 16, wherein the data captured at the UAS comprises two-dimensional (2D) video images of the object.
Zhu, Jiajun, Methods and systems for vehicle perception feedback to classify data representative of types of objects and to request feedback regarding such classifications.
Roy, Philippe; Yu, Jun; Linden, David S., Methods, apparatus and systems for enhanced synthetic vision and multi-sensor data fusion to improve operational capabilities of unmanned aerial vehicles.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.