Synthetic image automatic generation system and method thereof

• US 20090244071A1
• Filed: 07/14/2008
• Published: 10/01/2009
• Est. Priority Date: 08/09/2007
• Status: Abandoned Application

• Current Assignee: China Motor Corporation
• Original Assignee: China Motor Corporation
• Inventors: Kuo, Chien-Fu, Chang, Shao-Wen, Kuo, Chung An, Wang, Mao-Jiun, Lee, Hsu, Kuo, Feng Chou, Chen, Pei-Chao
• Application Number:
  US12/216,935
• Publication Number:
  US 20090244071A1
• Time in Patent Office: Days
• US Class Current: 345/473
• CPC Class Codes: G06F 40/55 Rule-based translation
• CPC Class Codes: G06T 13/40 of characters, e.g. humans, animals or virtual beings
• CPC Class Codes: G06T 2213/04 Animation description language

Abstract

Provided is a computer system and a computerized method to automatically generate the synthetic images that simulate the human activities in a particular environment. The program instructions are input in the form of the natural language. Particular columns are provided in the user interface to allow the user to select desired instruction elements from sets of limited candidates. The instruction elements form the program instructions. The system analyzes the program instructions to obtain the standard predetermined time evaluation codes of the instructions. Parameters not include in the input program instructions are generated automatically. Synthetic images are generated by using the input program instructions and the parameters obtained.

Claims

1. A synthetic image automatic generation system, comprising:

an instruction editor to provide an editing interface allowing the user to input natural language instructions defining a series of synthetic images that simulate the motion of a human body;

a simulation image generator to translate said input natural language instructions into motion instructions executable in particular digital human modeling computer program; and

a database, including a simulation image file storage section to store a plurality of simulation image files comprising a series of natural language instructions and a digital human modeling computer program storage section to store at least one digital human modeling computer program;

wherein said simulation image generator comprises a policy determination module and a simulation image generating module;

wherein said policy determination module generates according to attributes of elements of said natural language instructions parameters relating to said motion;

said parameters comprising at least one selected from the group consisted of position, posture and orientation of said human body and type of motion of said human body and wherein said simulation image generating module uses said natural language instructions and said parameters generated by said policy determination module to generate said motion instructions.

2. The synthetic image automatic generation system according to claim 1, wherein said natural language instruction comprises the descriptions of at least one operator, at least one operation and at least one object.

3. The synthetic image automatic generation system according to claim 2, wherein said natural language instruction further comprises descriptions in the conditions of said operation.

4. The synthetic image automatic generation system according to claim 2, wherein said editing interface provides selection columns allowing the user to select said operator, said operation and said object from a limited number of candidate operators, candidate operations and candidate objects.

5. The synthetic image automatic generation system according to claim 3, wherein said editing interface provides a selection column allowing the user to select said operator, said operation, said object and said conditions of operation from a limited number of candidate operators, candidate operations, candidate objects and candidate conditions.

6. The synthetic image automatic generation system according to claim 4, wherein said candidate operations comprise a plurality of operations to be performed by operators in the operation environment.

7. The synthetic image automatic generation system according to claim 6, wherein said candidate operations comprise a plurality of upper limb operations to be performed by operators in the operation environment.

8. The synthetic image automatic generation system according to claim 4, wherein said editing interface provides selection columns allowing the user to select in sequence said operator, said operation and said object from a limited number of candidate operators, candidate operations and candidate objects.

9. The synthetic image automatic generation system according to claim 5, wherein said editing interface provides selection columns allowing the user to select in sequence said operator, said operation, said object and said conditions from a limited number of candidate operators, candidate operations, candidate objects and candidate conditions.

10. The synthetic image automatic generation system according to claim 1, wherein said editing interface further provides an editing window to display at least one input natural language instruction and allows the user to modify contents of said natural language instruction and to change sequence of natural language instructions.

11. The synthetic image automatic generation system according to claim 1, wherein said policy determination module determines posture and position of said human body in particular time points according to a reach envelop model.

12. The synthetic image automatic generation system according to claim 11, wherein said policy determination module determines said posture and position according to a maximum comfortable reach line defined in said a reach envelop model.

13. The synthetic image automatic generation system according to claim 2, wherein said policy determination module determines posture and position of said human body in particular time points according to an operational policy selected from the group consisted of a concentration policy and a non-concentration policy and determined according to contents of said natural language instruction;

wherein said concentration policy requires said operator to perform said operation from a predetermined position facing a predetermined direction.

14. The synthetic image automatic generation system according to claim 2, wherein said policy determination module determines a path of motion of said human body in said operation according to an operational policy selected from the group consisted of a concentration policy and a non-concentration policy and determined according to contents of said natural language instruction;

wherein said concentration policy requires said operator to perform said operation from a predetermined position facing a predetermined direction.

15. The synthetic image automatic generation system according to claim 14, wherein path of motion of said human body of said concentration policy determined by said policy determination module comprises the motions of turning said human body, moving said human body forward and turning said human body to face an object.

16. The synthetic image automatic generation system according to claim 14, wherein path of motion of said human body of said non-concentration policy determined by said policy determination module comprises the motions of turning said human body and moving said human body to an object.

17. The synthetic image automatic generation system according to claim 1, further comprising an error detection module to detect errors contained in said natural language instruction.

18. A synthetic image automatic generation system, comprising:

an instruction editor to provide an editing interface allowing the user to input natural language instructions defining a series of synthetic images that simulate the motion of a human body; and

an instruction translator to convert said input natural language instructions into codes defined in a standard predetermined time evaluation system to be used in a digital human modeling computer program in the generation of said synthetic images;

wherein said instruction translator comprises a variables translator;

wherein said variable translator uses a look-up-table to convert said input natural language instructions into said standard predetermined time evaluation code according to attribute information contained in said natural language instruction.

19. The synthetic image automatic generation system according to claim 18, wherein said natural language instruction comprises the descriptions of at least one operator, at least one operation and at least one object.

20. The synthetic image automatic generation system according to claim 19, wherein said natural language instruction further comprises descriptions in the conditions of said operation.

21. The synthetic image automatic generation system according to claim 19, wherein said editing interface provides selection columns allowing the user to select said operator, said operation and said object from a limited number of candidate operators, candidate operations and candidate objects.

22. The synthetic image automatic generation system according to claim 20, wherein said editing interface provides a selection column allowing the user to select said operator, said operation, said object and said conditions of operation from a limited number of candidate operators, candidate operations, candidate objects and candidate conditions.

23. The synthetic image automatic generation system according to claim 21, wherein said candidate operations comprise a plurality of operations to be performed by operators in the operation environment.

24. The synthetic image automatic generation system according to claim 23, wherein said candidate operations comprise a plurality of upper limb operations to be performed by operators in the operation environment.

25. The synthetic image automatic generation system according to claim 21, wherein said editing interface provides selection columns allowing the user to select in sequence said operator, said operation and said object from a limited number of candidate operators, candidate operations and candidate objects.

26. The synthetic image automatic generation system according to claim 22, wherein said editing interface provides selection columns allowing the user to select in sequence said operator, said operation, said object and said conditions from a limited number of candidate operators, candidate operations, candidate objects and candidate conditions.

27. The synthetic image automatic generation system according to claim 18, wherein said editing interface further provides an editing window to display at least one input natural language instruction and allows the user to modify contents of said natural language instruction and to change sequence of natural language instructions.

28. The synthetic image automatic generation system according to claim 18, further comprising an error detection module to detect errors contained in said natural language instruction.

29. A method for the automatic generation of synthetic images, comprising the steps of:

obtaining a set of natural language instructions from an editing interface of a computer system;

each instruction comprising the motion elements of at least a human body description, at least an object description and at least an operation description

determining attributes of said motion elements contained in said natural language instruction;

generating motion parameters of said human body in said operation according to said obtained attribute information;

said motion parameters comprising information defining at least one selected from the group consisted of position, posture and orientation of said human body in said operation and speed and path of the motion of said human body in said operation;

generating motion instructions executable in a digital human modeling computer program according to said natural language instruction and said motion parameters; and

executing said motion instructions in said digital human modeling computer program to generate a series of synthetic images.

30. The method for the automatic generation of synthetic images according to claim 29, wherein said attributes of said motion elements are determined according to information provided in columns defined in said editing interface.

31. The method for the automatic generation of synthetic images according to claim 29, wherein position and posture of said human body is determined by using a reach envelop model.

32. The method for the automatic generation of synthetic images according to claim 31, wherein position and posture of said human body is determined according to a maximum comfortable reach line defined in said reach envelop model.

33. The method for the automatic generation of synthetic images according to claim 29, wherein posture and position of said human body in particular time points are determined according to an operational policy selected from the group consisted of a concentration policy and a non-concentration policy and determined according to contents of said natural language instruction;

wherein said concentration policy requires said operator to perform said operation from a predetermined position facing a predetermined direction.

34. The method for the automatic generation of synthetic images according to claim 29, wherein a path of motion of said human body in said operation is determined according to an operational policy selected from the group consisted of a concentration policy and a non-concentration policy and determined according to contents of said natural language instruction;

wherein said concentration policy requires said operator to perform said operation from a predetermined position facing a predetermined direction.

35. The method for the automatic generation of synthetic images according to claim 34, wherein said path of motion of said human body in said concentration policy comprises the motions of turning said human body, moving said human body forward and turning said human body to face an object.

36. The method for the automatic generation of synthetic images according to claim 34, wherein said path of motion of said human body in said non-concentration policy comprises the motions of turning said human body and moving said human body to an object.

37. The method for the automatic generation of synthetic images according to claim 34, further comprising the step of using said generated images as key-frames to generate intermediate images between each pair of two key-frames.

38. The method for the automatic generation of synthetic images according to claim 34, wherein said intermediate images are generated by interpolation.

39. The method for the automatic generation of synthetic images according to claim 29, further comprising the step of detecting errors in said input natural language instructions.

44. The method for the automatic generation of synthetic images according to claim 39, further comprising the step of detecting errors in said input natural language instructions.

40. A method for the automatic generation of synthetic images, comprising the steps of:

obtaining a set of natural language instructions from an editing interface of a computer system;

each instruction comprising the motion elements of at least a human body description, at least an object description and at least an operation description

obtaining in a look-up-table operation codes of said operation description in natural language instructions;

said look-up-table defining corresponding relations between said motion elements and codes defined in a standard predetermination time evaluation system;

obtaining from said look-up-table operational condition codes of said motion elements in said standard predetermined time evaluation system according attributes of said motion elements;

said attributes being determined according to attribute information defined in columns provided in said editing interface;

obtaining position information of said human body and said object from a digital human modeling computer program;

calculating to obtain distance information using said position information;

obtaining from said look-up-table distance codes corresponding said distance information in said standard predetermined time evaluation system;

outputting said operation codes, said operational condition codes and said distance codes as the standard predetermined time evaluation codes.

41. The method for the automatic generation of synthetic images according to claim 40, further comprising the step of executing said standard predetermined time evaluation codes in said digital human modeling computer program to generate synthetic images.

42. The method for the automatic generation of synthetic images according to claim 41, further comprising the step of using said generated images as key-frames to generate intermediate images between each pair of two key-frames.

43. The method for the automatic generation of synthetic images according to claim 42, wherein said intermediate images are generated by interpolation.