Image Tracing System and Method

Abstract

A method includes tagging, by at least one processor, one or more three-dimensional assets with a unique identifier and storing the one or more three-dimensional assets in a database, creating, by the at least one processor, a three-dimensional model based on the one or more three-dimensional assets and loading the three-dimensional model in a simulator, generating, by the at least one processor, a two-dimensional image that is a representation of the three-dimensional model in the simulator, the two-dimensional image comprising metadata that includes each unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image, and assigning, by the at least one processor, the two-dimensional image with a unique identifier and storing each unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image in metadata for the two-dimensional image.

What is claimed is: 
     
         1 . A method, comprising:
 tagging, by at least one processor, one or more three-dimensional assets with a global unique identifier and embedding each global unique identifier in a custom metadata field;   creating, by the at least one processor, a three-dimensional model based on the one or more three-dimensional assets;   generating, by the at least one processor, a two-dimensional image that is a representation of the three-dimensional model, the two-dimensional image comprising metadata that includes each global unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image; and   assigning, by the at least one processor, the two-dimensional image with a global unique identifier and storing each global unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image in metadata for the two-dimensional image.   
     
     
         2 . The method of  claim 1 , further comprising training a neural network using the generated two-dimensional image. 
     
     
         3 . The method of  claim 2 , further comprising generating one or more new two-dimensional images based on the generated two-dimensional image by transforming the generated two-dimensional image and training the neural network using the one or more new two-dimensional images. 
     
     
         4 . The method of  claim 2 , wherein the generated two-dimensional image is based on real world three-dimensional assets and synthetic three-dimensional assets. 
     
     
         5 . The method of  claim 3 , further comprising tracing each three-dimensional asset in the generated two-dimensional image back to an original simulation event and to original source assets. 
     
     
         6 . The method of  claim 1 , further comprising generating the one or more three-dimensional assets using an asset manager that receives input from a user using a web-based graphical user interface (GUI), the asset manager receiving the global unique identifier for each of the one or more three-dimensional assets. 
     
     
         7 . The method of  claim 6 , further comprising generating a simulation using the one or more three-dimensional assets that receives the input from the user using the web-based GUI. 
     
     
         8 . The method of  claim 7 , further comprising defining at least one event in the simulation, defining weather, defining a time of day, and placing at least one camera in the simulation. 
     
     
         9 . The method of  claim 1 , further comprising generating a plurality of two-dimensional images based on the generated two-dimensional image and using the plurality of two-dimensional images as output for a neural network training framework. 
     
     
         10 . The method of  claim 1 , wherein the two-dimensional image comprises a Portable Network Graphics (PNG) file. 
     
     
         11 . The method of  claim 1 , further comprising storing the one or more three-dimensional assets in a database. 
     
     
         12 . The method of  claim 1 , wherein the metadata comprises an exchangeable image file format (Exif) field that is a Javascript Object Notation (JSON) string that comprises an id field for the global unique identifier, a pass_number field, an initial_sim_time field, a sim_event field, a modified field, a source_immediate field, a source_ultimate field, a transform field, and a tool_parameters field. 
     
     
         13 . A system comprising:
 a memory storing computer-readable instructions; and   at least one processor to execute the instructions to:   tag one or more three-dimensional assets with a global unique identifier and embed each global unique identifier in a custom metadata field;   create a three-dimensional model based on the one or more three-dimensional assets and load the three-dimensional model;   generate a two-dimensional image that is a representation of the three-dimensional model, the two-dimensional image comprising metadata that includes each global unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image; and   assign the two-dimensional image with a global unique identifier and store each global unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image in metadata for the two-dimensional image.   
     
     
         14 . The system of  claim 13 , the at least one processor further to execute the instructions to store the one or more three-dimensional assets in a database. 
     
     
         15 . The system of  claim 13 , wherein the metadata comprises an exchangeable image file format (Exif) field that is a Javascript Object Notation (JSON) string that comprises an id field for the global unique identifier, a pass_number field, an initial_sim_time field, a sim_event field, a modified field, a source_immediate field, a source_ultimate field, a transform field, and a tool_parameters field. 
     
     
         16 . The system of  claim 13 , the at least one processor further to execute the instructions to train a neural network using the generated two-dimensional image. 
     
     
         17 . The system of  claim 16 , the at least one processor further to execute the instructions to generate one or more new two-dimensional images based on the generated two-dimensional image by transforming the generated two-dimensional image and training the neural network using the one or more new two-dimensional images. 
     
     
         18 . The system of  claim 16 , wherein the generated two-dimensional image is based on real world three-dimensional assets and synthetic three-dimensional assets. 
     
     
         19 . The system of  claim 18 , the at least one processor further to execute the instructions to trace each three-dimensional asset in the two-dimensional image back to an original simulation event and to original source assets. 
     
     
         20 . A non-transitory computer-readable storage medium comprising instructions stored thereon that, when executed by a computing device cause the computing device to perform operations, the operations comprising:
 tagging one or more three-dimensional assets with a global unique identifier and embedding each global unique identifier in a custom metadata field;   creating a three-dimensional model based on the one or more three-dimensional assets and loading the three-dimensional model;   generating a two-dimensional image that is a representation of the three-dimensional model, the two-dimensional image comprising metadata that includes each global unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image; and   assigning the two-dimensional image with a global unique identifier and storing global each unique identifier for each three-dimensional asset of the three-dimensional model displayed in the two-dimensional image in metadata for the two-dimensional image.