Apparatus and method for editing data and program

Abstract

A flexible data editing scheme to change and modify an intermediate representation or conditional information for a portion of to-be-edited data is disclosed. One aspect of the present disclosure relates to a data editing apparatus, comprising: one or more memories; and one or more processors configured to receive a change indication to change at least a first data area of first data; generate second data by using one or more generative models and an intermediate representation for the first data area; and replace the first data area of the first data with the second data to generate third data.

What is claimed is: 
     
         1 . A data editing apparatus, comprising:
 one or more memories; and   one or more processors configured to:
 receive a change indication to change a first data area of first data; 
 generate second data by using one or more generative models and an intermediate representation for the first data area; and 
 replace the first data area of the first data with the second data to generate third data. 
   
     
     
         2 . The data editing apparatus as claimed in  claim 1 , wherein the one or more generative models are selected from a plurality of generative models based on the change indication. 
     
     
         3 . The data editing apparatus as claimed in  claim 1 , wherein the processors are configured to:
 modify the intermediate representation for the first data area by using an intermediate representation obtained from fourth data; and   generate the second data by using the modified intermediate representation.   
     
     
         4 . The data editing apparatus as claimed in  claim 1 , wherein the processors are configured to generate the second data by using conditional information based on the change indication. 
     
     
         5 . The data editing apparatus as claimed in  claim 1 , wherein the first data represents an image, and the processors are further configured to detect at least the first data area including an object from the first data. 
     
     
         6 . The data editing apparatus as claimed in  claim 5 , wherein the processors are further configured to detect class information indicative of a class of the object. 
     
     
         7 . The data editing apparatus as claimed in  claim 1 , wherein the one or more generative models generate the second data gradually, and the processors are configured to change at least one of an intermediate representation or conditional information for each phase or depending on phases and input the changed intermediate representation or the changed conditional information to the one or more generative models to generate the second data. 
     
     
         8 . The data editing apparatus as claimed in  claim 7 , wherein the one or more generative models are implemented with a neural network formed of a plurality of layers and generate the second data with the respective layers gradually. 
     
     
         9 . The data editing apparatus as claimed in  claim 1 , wherein the first data represents an image, and the processors are configured to receive the change indication to change conditional information for a second data area in the first data area including an object in the image, the second data area including an image area for a portion of the object. 
     
     
         10 . A data editing method, comprising:
 receiving, by one or more processors, a change indication to change at least a first data area of first data;   generating, by the one or more processors, second data by using one or more generative models and an intermediate representation for the first data area; and   replacing, by the one or more processors, the first data area of the first data with the second data to generate third data.   
     
     
         11 . The data editing method as claimed in  claim 10 , wherein the one or more generative models are selected from a plurality of generative models based on the change indication. 
     
     
         12 . The data editing method as claimed in  claim 10 , further comprising:
 modifying, by the one or more processors, the intermediate representation for the first data area by using an intermediate representation obtained from fourth data; and   generating, by the one or more processors, the second data by using the modified intermediate representation.   
     
     
         13 . The data editing method as claimed in  claim 10 , further comprising:
 generating, by the one or more processors, the second data by using conditional information based on the change indication.   
     
     
         14 . The data editing method as claimed in  claim 10 , wherein the first data represents an image, and the method further comprising:
 detecting, by the one or more processors, at least the first data area including an object from the first data.   
     
     
         15 . The data editing method as claimed in  claim 14 , further comprising:
 detecting, by the one or more processors, class information indicative of a class of the object.   
     
     
         16 . The data editing method as claimed in  claim 10 , wherein the one or more generative models generate the second data gradually, and the method further comprising:
 changing, by the one or more processors, at least one of an intermediate representation or conditional information for each phase or depending on phases; and   inputting, by the one or more processors, the changed intermediate representation or the changed conditional information to the one or more generative models to generate the second data.   
     
     
         17 . The data editing method as claimed in  claim 16 , wherein the one or more generative models are implemented with a neural network formed of a plurality of layers and generate the second data with the respective layers gradually. 
     
     
         18 . The data editing method as claimed in  claim 10 , wherein the first data represents an image, and the method further comprising:
 receiving, by the one or more processors, the change indication to change conditional information for a second data area in the first data area including an object in the image, the second data area including an image area for a portion of the object.   
     
     
         19 . A storage medium for storing instructions for causing one or more computers to:
 receive a change indication to change at least a first data area of first data;   generate second data by using one or more generative models and an intermediate representation for the first data area; and   replace the first data area of the first data with the second data to generate third data.   
     
     
         20 . The storage medium as claimed in  claim 19 , wherein the instructions further cause the one or more computers to:
 modify the intermediate representation for the first data area by using an intermediate representation obtained from fourth data; and   generate the second data by using the modified intermediate representation.