A SYSTEM AND A METHOD FOR CAPTURING AND GENERATING 3D IMAGE Technical Field

Present invention is related to a method and system for generating a 3 dimensional image of a target using a single camera. Prior Art

3 dimensional image of a target element or a scene is usually generated by using plurality of (at least two) image recorders (cameras), wherein each image recorder records an image of said target element or scene at a different angle. By combining the images recorded by said image recorders, depth information for the target element or scene is generated. Thus, by using the depth information, a 3D image of the target element or scene is able to be generated.

In order to generate depth information, most of the known techniques require at least two image recorders. However, using two image recorders is an expensive solution. Moreover, using two image recorders on a single device (for example a portable device like a mobile phone) causes heating problem on said device.

In order to overcome said problems, systems that generate 3D images using single image recorder are proposed. Article of Thomas Lu and Tien-Hsin Chao "A single-camera system captures high-resolution 3D images in one shot" (DOI: 10.1 1 17/2.120061 1.0303) discloses a system and method for capturing a 3D image using a single camera. According to said article, an optical system is attached to the camera in order to capture two different images from different angles. Then, said images are combined so as to generate a 3D image. However, according to said article, since 2 different images are generated at the same time, resolution of each image is half of the resolution of the camera. In other words, according to said article, resolution of the generated 3D image is less than the actual resolution of the camera. Moreover, since said method requires an rare found optical system, also the dimensions of the camera increases. Therefore, said method may not be used with mobile devices, sizes of which are limited.

Brief Description of the Invention

According to the present invention, a method and a system for generating a 3 dimensional image of a target using a single camera are provided. Said system comprises, at least one light source, intensity of which is able to be adjusted; at least one image recorder, which captures images of a target comprising plurality of obstacles; and at least one control unit (not shown in figures), which controls said light source by increasing or decreasing its intensity within a time period, which controls said image recorder so as to capture plurality of images of said target within said time period, and which determines the depth of said obstacles so as to capture and generate a 3 dimensional image by comparing the illumination level change of the obstacles between captured images within said time period. Said method comprises the steps of, increasing or decreasing the intensity of said light source within a time period; capturing the images of said target via said image recorder within said time period; and determining the depth of said obstacles, so as to capture and generate a 3 dimensional image, by comparing the illumination level change of the obstacles between captured images within said time period via at least one control unit.

According to the present application, since one image recorder, intensity adjustable light source and a control unit is enough for capturing and generating a 3D image, conventional portable devices with cameras, such as mobile phones/smart phones, can be used for capturing and generating 3D images. Therefore, without changing the hardware of a mobile phone/smart phone, 3D images are able to be captured and generated easily.

Object of the Invention The main object of the present invention is to provide a system and method for capturing and generating a 3 dimensional image of a target using a single camera.

Another aim of the present invention is to provide a system and method for capturing and generating a 3 dimensional image which can be easily implemented on portable devices. Another aim of the present invention is to provide a system and method for capturing and generating a 3 dimensional video of a target using a single camera. Another aim of the present invention is to provide a cheap and reliable system and method for capturing and generating a 3 dimensional image and video.

Description of Drawings Figure 1 shows an exemplary embodiment of the 3 dimensional image capture and generation system of the present invention.

Figure 2 shows another exemplary embodiment of the 3 dimensional image capture and generation system of the present invention.

Figure 3 shows an intensity of a light source versus time graph of an exemplary embodiment of the present invention.

Figure 4 shows another intensity of a light source versus time graph of an exemplary embodiment of the present invention.

The references in the figures may possess following meanings;

Light source (1)

Image recorder (2)

Illumination range (3)

Obstacle (4)

Detailed Description of the invention

3 dimensional images contain image data and depth data. In the conventional methods, in order to capture and generate a 3D image of a target (an object or a scene), images having different viewpoints are used. Therefore, conventional methods require plurality of image recorders (cameras) in order to capture and generate a 3D image. According to the present invention, a method and system for capturing and generating a 3 dimensional image of a target using a single camera are provided. 3 dimensional image capture and generation system of the present invention, exemplary embodiments of which are given in figures 1-4, comprises at least one light source (1), intensity of which is able to be adjusted; at least one image recorder (2) (such as a camera), which captures images (2 dimensional images) of a target comprising plurality of obstacles (4); and at least one control unit (not shown in figures), which controls said light source (1) by increasing or decreasing its intensity within a time period, which controls said image recorder (2) so as to capture plurality of images of said target within said time period, and which determines the depth of said obstacles (4) so as to capture and generate a 3 dimensional image by comparing the illumination level change of the obstacles (4) between captured images within said time period.

3 dimensional image capture and generation method of the present invention comprises the steps of, increasing or decreasing the intensity of at least one light source (1) within a time period, wherein said light source (1) illuminates a target comprising plurality of obstacles (4); capturing the images of said target via at least one image recorder (2) within said time period; determining the depth of said obstacles (4), so as to capture and generate a 3 dimensional image, by comparing the illumination level change of the obstacles (4) between captured images within said time period via at least one control unit. In a preferred embodiment, by using the determined depth of the obstacles (4), a saliency map, which is sued in 3 dimensional image applications, is generated.

An exemplary embodiment of the present invention, which is given in figure , intensity of the light source (1) is increased within a time period. At the start of said time period, 1a time, intensity of light source (1) is so low that illumination range (3) of the light source (1) is only able to illuminate the obstacle (4) closest to the light source (1). During that time, image captured by the image recorder (2) shows that only the obstacle (4) closest to the light source (1) is illuminated. When the intensity of the light source (1) is increased, 1b time, illumination range (3) of the light source (1) reaches the obstacle (4), which is second closest to the light source (1). During 1 b time, image captured by the image recorder (2) shows that only the two obstacles (4) that are closest to the light source (1) are illuminated. By comparing the illumination level change of two obstacles (4) that are closest to the light source (1) between the images captured at 1a time and 1 b time, distances between the light source (1) and said obstacles (4) are able to be calculated. By further increasing the intensity of the light source (1) at 1c and d times and comparing the illumination level change of the obstacles (4) during those times, distances between the light source (1) and said obstacles (4) are able to be calculated. By using said distance information and distance between the light source (1) and image recorder (2), distances between the image recorder (2) and said obstacles (4) (which are the depths of said obstacles (4)) are calculated. This embodiment is preferred in relatively dark environment because illumination level increase of the obstacles (4) over time becomes apparent. Moreover, When the highest illumination range (3) is achieved (as shown in figure 1d and 2a), it is also possible to determine the depth order of the illuminated obstacles (4) according to the colour change differentiation level of each obstacle (4). The closest obstacle (4) will be brighter and the colour differentiation will be the highest and vice versa.

An alternative exemplary embodiment of the present invention is given in figure 2. In this embodiment, intensity of the light source (1) is decreased within a time period. Similar to previous embodiment, by using illumination level change of the obstacles (4) over different times (2a-2d times), depth information for each of the obstacles (4) are able to be calculated. In this embodiment, illumination level of the obstacles (4) decreases over time. Therefore, this embodiment is preferred in relatively bright environments. In a preferred embodiment of the present invention, intensity of the light source (1) changes (increases or decreases) stepwisely as shown in figure 3. Steps of intensity may change discretely (in other words intensity of the light source (1) becomes 0 for a short time between two steps) or may change continuously (in other words intensity of the light source (1) increases or decreases as stair steps). In these embodiments, width of each step (time gap of each intensity level) is preferably equal to or greater than the shutter speed (exposure time) of the image recorder (2), so that intensity of the light source (1) does not change within an exposure time. Therefore, illumination levels of the obstacles (4) are able to be determined more accurately. In another preferred embodiment of the present application, intensity of the light source (1) changes in continuous manner as shown in figure 4.

In another embodiment of the present application, colour (wavelength) of the light source (1) is able to be adjusted by said control unit. Since different coloured obstacles (4) absorb different coloured lights, illumination level change of an obstacle (4) may not be apparent if it is illuminated with a specific coloured light source (1). Therefore, according to this embodiment, if illumination level change of an obstacle (4) is not apparent within a time period, colour of the light source (1) is changed and steps of increasing or decreasing the intensity of at least one light source (1); capturing the images of said target via at least one image recorder (2) and determining the depth of said obstacles (4) are repeated for the new colour. Therefore, depth information of the obstacles (4) are determined more accurately. In another embodiment, light intensity of the light source (1) outspreads in semispherical form. In this embodiment, depth of the obstacles are able to be calculated in a polar coordinate system.

In another preferred embodiment of the present application, 3 dimensional image capture and generation system is used for capturing and generating 3 dimensional videos. In this embodiment, firstly depth of different obstacles (4) are determined according to above mentioned applications and initial sizes of the obstacles (4) are determined within captured images (size of the obstacles (4) are determined in terms of pixels). Then, a video is recorded using said image recorder (2) or another video recording device. During said recording, depth of the obstacles (4) may change. In order to determine said depth change, relative size changes of the obstacles (4) are measured (again in terms of pixels). Since said relative size change is related with the depth change, new depth of the obstacles (4) are able to be calculated. In another preferred embodiment of the present application, after the depth of the obstacles (4) are determined, said depth information is used for distinguish a foreground object from a background object. Thus, it is possible to change a background object to another captured image or video. In other words, according to the system and method of the present application, greenbox (or bluebox or same purpose) applications are able to be used without a green (or blue) screen.

According to the present invention, one image recorder (2) is enough for determining depth information and capturing and generating a 3D image. However, in alternative embodiments of the present application, plurality of image recorders (2) may be used. In the conventional application, wherein at least two cameras are used for generating a 3D image, viewpoint of the cameras are different. Therefore, if one segment of a target is seen by only one camera, intrusions/extrusions on said segment cannot be identified. The embodiment, wherein plurality of image recorders (2) are used, overcomes said problem. Therefore, according to the present application, more accurate depth information and more accurate 3D image are able to be captured and generated.

According to the present application, since one image recorder (2), intensity adjustable light source (1) and a control unit is enough for generating a 3D image, conventional portable devices with cameras, such as mobile phones/smart phones, can be used for capturing and generating 3D images. Therefore, without changing the hardware of a mobile phone/smart phone, 3D images are able to be captured and generated easily.