TWI439962B - Intuitive image depth generation system and image depth generation method - Google Patents

Description

Intuitive depth image generation system and depth image generation method

The present invention relates to an image generation system and an image generation method, and more particularly to a depth image generation system and a depth image generation method.

Due to the advancement of technology, the development of today's electronic products is not a priority to improve the performance and quality of products. Among them, regarding the design of the display device, in addition to providing a clearer and more comfortable display screen for the user, a third-dimension (3D) display device has been proposed to provide the user with a different visual experience.

At this stage, due to the lack of 3D image content that can be used in 3D display devices, the technology from 2D image to 3D image has become a problem that the industry and academia have jointly discussed and solved in recent years. The existing 3D image generation system generates a 3D image format supported by the 3D display device by inputting a specific reference value by the user to generate a depth result of the 2D image, and then using the original 2D image and the depth result. However, such an image generation method is too focused on a given reference value, and therefore, it takes a relatively long production time, and the user cannot easily perform 3D image generation operations.

Therefore, how to provide a depth image generation system and a depth image generation method, which can make the user easy to operate, thereby improving the speed and quality of image generation, has become one of the important topics.

In view of the above problems, an object of the present invention is to provide a depth image generation system and a depth image generation method that can facilitate operation by a user and thereby improve the speed and quality of image generation.

To achieve the above object, a depth image generating system according to the present invention is used in combination with a computer device, which comprises an object capturing module and an object depth generating module. The object capture module selects an object of an image data and generates an outline image of the object. The object depth generation module receives the contour map of the object and generates the depth value of the object.

In an embodiment of the invention, the object capture module selects the object according to a first instruction input by a user.

In an embodiment of the invention, the object depth generation module generates a depth value of the object according to a second instruction or a corresponding relationship between the object and a background depth value of the image data, wherein the second instruction includes a position reference of the specified object. Point and or one of the gradient reference points of the object.

In order to achieve the above object, a depth image generating method for a depth image generating system according to the present invention, wherein the depth image generating system is used in combination with a computer device, and the depth image generating system has an object capturing module and an object depth generating module. The method for generating a depth image includes the following steps: selecting an object of an image data by the object capturing module, and generating a contour map of the object; and generating a module by the object depth generating module, receiving the contour map of the object, and generating the depth of the object. value.

In an embodiment of the invention, the object capture module selects an object according to a first instruction input by a user.

In an embodiment of the invention, the object depth generation module generates the depth value of the object according to a second instruction or a corresponding relationship between the object and a background depth value of the image data.

To achieve the above objective, the present invention also discloses a computer program product for storing deep image generation. When a computer device is loaded into the computer program and executed, the depth image generation method described above can be completed.

According to the present invention, a depth image generating system and a depth image generating method according to the present invention select an object of an image data by using the object capturing module, and generate a contour map of the object to make the object depth generated. The module receives the contour map of the object and produces the depth value of the object. Therefore, the user can operate easily, thereby improving the speed and quality of image generation.

Hereinafter, a depth image generating system and a depth image generating method according to a preferred embodiment of the present invention will be described with reference to related drawings.

Please refer to FIG. 1. FIG. 1 is a schematic diagram of a depth image generating system and a computer device according to a preferred embodiment of the present invention. The depth image generation system 1 is used in conjunction with the computer device 2, and the depth image generation system 1 includes an object capture module 11 and an object depth generation module 12.

The object capture module 11 is configured to select one of the image data displayed by the display 21 of the computer device 2, and generate a contour map of the selected object. In this embodiment, the object capture module 11 selects an object in the image data according to a first command C1 input by the user. The first instruction C1 may be an instruction directly designating an object to be selected, or an instruction of an initial contour of a given object.

In practical use, after receiving the first command C1, the object capturing module 11 can adopt different processing manners to generate a contour map of the specified object, for example, in a general state, using Sobel edge detection. Sobel edge detection produces contour maps of objects, uses Canny Edge detection in images that require special emphasis on straight lines, and uses homogenous edge detection when images are blurry. Detection) to expand the difference between the objects of the image data to produce a contour map of the object.

In addition, it is worth mentioning that the object capture module 11 can determine the state of the image data and the specified object, such as brightness, contrast, and resolution, according to one of the preset criteria, and automatically select the desired image. The detection method, or the user selects the required detection method to generate the contour map of the object.

The object depth generation module 12 receives the contour map of the object and generates a depth value of the object. In this embodiment, the object depth generation module 12 generates the depth value of the object according to a second command C2, and displays the contour map and the depth value of the object on the display 21 of the computer device 2.

In practice, the second command C2 can be an instruction input by the user, and it includes a position reference point of one of the specified objects and a gradient reference point of the object. Wherein, the position reference point is located on the edge of the contour map of the object or in the contour map, and the object depth generation module 12 generates the depth value of one row of the position reference point based on the specified position reference point, and the line The depth values are all the same, but a similar depth value.

In addition, the gradient reference point may be the lowest point or the highest point of each line in the image data, and the object depth generation module 12 uses the specified gradient reference point as a starting point or an end point to generate a gradient reference point. The depth value of a row, and the depth value of this row is a gradient depth value. Wherein, the gradient depth value is such that the depth value of the entire row has a change according to a function. The aforementioned functions are, for example, linear functions, exponential functions, logarithmic functions or other non-linear functions.

In actual use, the depth image generation system 1 may further include an image data storage module (not shown). The image data storage module stores a plurality of background images and a complex array setting parameter. The object depth generating module 12 can determine the corresponding relationship between the object and the background depth value of the image data by comparing and calculating various information stored in the image data storage module, thereby generating the object capturing module 11 The depth value of the selected object.

In addition, it is worth mentioning that the object depth generation module 12 can also generate the depth values of the remaining objects in the same image data according to the depth values of the objects already generated in the image data.

Therefore, with the above-described depth image generation system 1, the user can process the image data in a more intuitive and simple manner by inputting a specific reference value without using a specific reference value.

Next, please refer to FIG. 2. FIG. 2 is a depth image generating system 3 according to a preferred embodiment of the present invention. The depth image generation system 3 differs from the image generation system 1 in that the depth image generation system 3 further includes an image display module 13. The image display module 13 is configured to display the contour map of the object and the depth value of the object, so that the user can observe various patterns in each processing stage and determine the subsequent operation mode. In addition, the image display module 13 can also generate a depth map of the image data according to the depth values of all the objects of the image data, and generate a 3D image by using the original image data and the depth map of the image data.

Next, please refer to FIG. 3 , which will be described below with reference to the flowchart of FIG. 3 and FIG. 1 to illustrate a depth image generating method according to a preferred embodiment of the present invention, which is related to, for example, the depth image generating system 1 described above. For use with. The steps of the depth image generation method include S01 to S02.

In step S01, an object image is selected by the object capturing module 11 to generate an outline image of the object. In this embodiment, the object capturing module 11 selects an object in the image data according to the first instruction input by the user, and generates a contour map of the selected object.

Step S02 is performed by the object depth generation module 12, which receives the contour map of the object and generates the depth value of the object. In this embodiment, the object depth generation module 12 generates the depth value of the object according to the second instruction or the corresponding relationship between the object and the background depth value of the image data.

In practice, steps S01-S02 may be repeatedly performed to generate depth values of all objects in the image data, thereby generating a depth map of the image data. Among them, all the aforementioned objects include the foreground and background of the image data. In addition, the user can select one object at a time to separately generate the contour map and the depth value, or select multiple objects at a time to simultaneously generate the contour map and the depth value.

Next, in order to more clearly explain the depth image generation system and the depth image generation method according to the present invention, please refer to the schematic diagrams of image generation of FIGS. 4A to 4D.

As shown in FIG. 4A, the computer device displays an image data on its display. It should be noted that the present embodiment is based on the fact that the image data has characters, vehicles, buildings, and the sky as an example, and is not limited thereto.

As shown in FIG. 4B, when the first instruction specifies that the object to be selected is a character and a vehicle in the image data, the object capturing module selects the character and the vehicle according to the first instruction, and generates the selected object. Outline map of people and vehicles.

As shown in FIG. 4C, when the first instruction sequentially specifies that the object to be selected is the building and the sky in the image data, the object capturing module selects the specified object according to the first instruction. In addition, in the present embodiment, the second command specifies the position reference point of the person, the vehicle, and the sky, respectively, and the gradation reference point of the building. Therefore, the object depth generation module will generate the depth value of each object according to the second instruction.

As shown in FIG. 4D, after completing the depth values of all the objects of the image data, the image display module will generate a depth map of the image data. In addition, the image display module can generate a 3D image by using the original image data and the depth map of the image data.

It should be noted that the present invention also discloses a computer program product for storing image generation. When the computer device is loaded into a computer program and executed, the above-described depth image generation method can be completed. The computer program can store the code in a recording medium or a code that can be transmitted through the network, so the computer device can directly read the recording medium or download the code through the network to load the code, and further execute the program. The above method of generating a depth image is completed.

In summary, according to the depth image generation system and the depth image generation method of the present invention, an object image is selected by using the object capture module, and an outline image of the object is generated to make the object depth generated. The module receives the contour map of the object and produces the depth value of the object. Therefore, the user can operate easily, thereby improving the speed and quality of image generation.

The above is intended to be illustrative only and not limiting. Any equivalent modifications or alterations to the spirit and scope of the invention are intended to be included in the scope of the appended claims.

1, 3. . . Depth image generation system

11. . . Object capture module

12. . . Object depth generation module

13. . . Image display module

2. . . Computer device

twenty one. . . monitor

C1. . . First instruction

C2. . . Second instruction

S01～S02. . . Steps of the depth image generation method

1 is a schematic diagram of a depth image generating system in accordance with a preferred embodiment of the present invention;

2 is a schematic diagram of a depth image generation system in accordance with a preferred embodiment of the present invention;

3 is a flowchart of a depth image generation method of a depth image generation system according to a preferred embodiment of the present invention;

4A-4D are schematic diagrams showing image generation of a depth image generation method according to a preferred embodiment of the present invention.

1. . . Depth image generation system

11. . . Object capture module

12. . . Object depth generation module

2. . . Computer device

twenty one. . . monitor

C1. . . First instruction

C2. . . Second instruction

Claims (10)

A depth image generating system is used in combination with a computer device. The depth image generating system comprises: an object capturing module, selecting an object of an image data, and generating a contour map of the object; and an object depth generating module. Receiving a contour map of the object and generating a depth value of the object; and an image display module, generating a depth map of the image data according to the depth value of all the objects of the image data, and using the image data and the depth map Generate a 3D image. The depth image generating system of claim 1, wherein the object capturing module selects the object according to a first instruction input by a user, and the first instruction is an instruction directly specifying the object. , or an instruction for the initial outline of the object. The depth image generating system of claim 1, wherein the image display module further displays a contour map of the object and a depth value of the object. The depth image generation system of claim 1, wherein the object capture module generates a contour map of the object by a Sobel edge detection method, a Canney edge detection method, or a homogenous edge detection method. The depth image generation system of claim 1, wherein the object depth generation module generates a depth value of the object according to a second instruction or a corresponding relationship between the object and a background depth value of the image data. And the second instruction includes specifying a position reference of the object Point and or a gradient reference point of the object. The depth image generating system of claim 1, wherein the depth value of the object is a progressive depth value or a similar depth value. A depth image generation method is applied to a depth image generation system, wherein the depth image generation system is used in combination with a computer device, and the depth image generation system has an object capture module, an object depth generation module, and an image. a display module, the depth image generation method includes the following steps: selecting, by the object capture module, an object of an image data, and generating a contour map of the object; and generating an outline of the object by the object depth generation module And generating a depth map of the object according to the depth value of all the objects of the image data, and generating a 3D image by using the image data and the depth map. The depth image generating method according to claim 7, wherein the object capturing module selects the object according to a first instruction input by a user, and the first instruction is an instruction directly specifying the object. , or an instruction for the initial outline of the object. The method for generating a depth image according to the seventh aspect of the invention, wherein the object depth generation module generates a depth value of the object according to a second instruction or a corresponding relationship between the object and a background depth value of the image data. And the second instruction includes a position reference point specifying one of the objects and a gradient reference point of the object. A computer program product for storing deep image generation, when a computer device is loaded into the computer program and executed, the depth image generation method described in claim 7 can be completed.